2020.07.17.txt

==========New Papers==========
1, TITLE:       SSN: Soft Shadow Network for Image Compositing
http://arxiv.org/abs/2007.08211
AUTHORS:        Yichen Sheng ; Jianming Zhang ; Bedrich Benes
COMMENTS:       11 pages, 10 figures
HIGHLIGHT:      We introduce a Soft Shadow Network to generate convincing soft shadows for 2D object cutouts automatically.

2, TITLE:       Tracking Passengers and Baggage Items using Multi-camera Systems at Security Checkpoints
http://arxiv.org/abs/2007.07924
AUTHORS:        Abubakar Siddique ; Henry Medeiros
COMMENTS:       14 pages, 11 figures
HIGHLIGHT:      We introduce a novel tracking-by-detection framework to track multiple objects in overhead camera videos for airport checkpoint security scenarios where targets correspond to passengers and their baggage items.

3, TITLE:       SLK-NER: Exploiting Second-order Lexicon Knowledge for Chinese NER
http://arxiv.org/abs/2007.08416
AUTHORS:        Dou Hu ; Lingwei Wei
COMMENTS:       5 pages, The work is accepted by SEKE2020
HIGHLIGHT:      Based on these, we propose a SLK-based model with a novel strategy to integrate the above lexicon knowledge.

4, TITLE:       Filter Style Transfer between Photos
http://arxiv.org/abs/2007.07925
AUTHORS:        Jonghwa Yim ; Jisung Yoo ; Won-joon Do ; Beomsu Kim ; Jihwan Choe
COMMENTS:       ECCV (Spotlight) 2020
HIGHLIGHT:      In this paper, we introduce a new concept of style transfer, Filter Style Transfer (FST).

5, TITLE:       A Refined Deep Learning Architecture for Diabetic Foot Ulcers Detection
http://arxiv.org/abs/2007.07922
AUTHORS:        Manu Goyal ; Saeed Hassanpour
COMMENTS:       8 Pages and DFUC Challenge
HIGHLIGHT:      In this paper, we propose using deep learning methods (EfficientDet Architectures) for the detection of DFU in the DFUC2020 challenge dataset, which consists of 4,500 DFU images.

6, TITLE:       Image De-Quantization Using Generative Models as Priors
http://arxiv.org/abs/2007.07923
AUTHORS:        Kalliopi Basioti ; George V. Moustakides
HIGHLIGHT:      Our goal in this work is to develop a de-quantization mechanism through a rigorous mathematical analysis which is based on the classical statistical estimation theory.

7, TITLE:       Appearance-Preserving 3D Convolution for Video-based Person Re-identification
http://arxiv.org/abs/2007.08434
AUTHORS:        Xinqian Gu ; Hong Chang ; Bingpeng Ma ; Hongkai Zhang ; Xilin Chen
COMMENTS:       Accepted by ECCV2020 (Oral)
HIGHLIGHT:      To address this problem, we propose AppearancePreserving 3D Convolution (AP3D), which is composed of two components: an Appearance-Preserving Module (APM) and a 3D convolution kernel.

8, TITLE:       Meta-Gradient Reinforcement Learning with an Objective Discovered Online
http://arxiv.org/abs/2007.08433
AUTHORS:        Zhongwen Xu ; Hado van Hasselt ; Matteo Hessel ; Junhyuk Oh ; Satinder Singh ; David Silver
HIGHLIGHT:      In this work, we propose an algorithm based on meta-gradient descent that discovers its own objective, flexibly parameterised by a deep neural network, solely from interactive experience with its environment.

9, TITLE:       ClassMix: Segmentation-Based Data Augmentation for Semi-Supervised Learning
http://arxiv.org/abs/2007.07936
AUTHORS:        Viktor Olsson ; Wilhelm Tranheden ; Juliano Pinto ; Lennart Svensson
HIGHLIGHT:      We propose a novel data augmentation mechanism called ClassMix, which generates augmentations by mixing unlabelled samples, by leveraging on the network's predictions for respecting object boundaries.

10, TITLE:      Improving rigid 3D calibration for robotic surgery
http://arxiv.org/abs/2007.08427
AUTHORS:        Andrea Roberti ; Nicola Piccinelli ; Daniele Meli ; Riccardo Muradore ; Paolo Fiorini
COMMENTS:       Submitted to the special issue of IEEE Transactions on Medical Robotics and Bionics 2020
HIGHLIGHT:      In this paper, we propose a novel calibration technique for a surgical scenario with da Vinci robot.

11, TITLE:      Investigating Pretrained Language Models for Graph-to-Text Generation
http://arxiv.org/abs/2007.08426
AUTHORS:        Leonardo F. R. Ribeiro ; Martin Schmitt ; Hinrich Schütze ; Iryna Gurevych
HIGHLIGHT:      In this paper, we aim to investigate the impact of large PLMs in graph-to-text generation.

12, TITLE:      Attention-Based Query Expansion Learning
http://arxiv.org/abs/2007.08019
AUTHORS:        Albert Gordo ; Filip Radenovic ; Tamara Berg
COMMENTS:       Accepted for publication at ECCV2020
HIGHLIGHT:      In this paper we propose a more principled framework to query expansion, where one trains, in a discriminative manner, a model that learns how images should be aggregated to form the expanded query.

13, TITLE:      An Empirical Study on the Robustness of NAS based Architectures
http://arxiv.org/abs/2007.08428
AUTHORS:        Chaitanya Devaguptapu ; Devansh Agarwal ; Gaurav Mittal ; Vineeth N Balasubramanian
HIGHLIGHT:      In this work, we study the adversarial robustness of existing NAS architectures, comparing it with state-of-the-art handcrafted architectures, and provide reasons for why it is essential.

14, TITLE:      Active Visual Information Gathering for Vision-Language Navigation
http://arxiv.org/abs/2007.08037
AUTHORS:        Hanqing Wang ; Wenguan Wang ; Tianmin Shu ; Wei Liang ; Jianbing Shen
COMMENTS:       ECCV2020; website: https://github.com/HanqingWangAI/Active_VLN
HIGHLIGHT:      To achieve this, we propose an end-to-end framework for learning an exploration policy that decides i) when and where to explore, ii) what information is worth gathering during exploration, and iii) how to adjust the navigation decision after the exploration.

15, TITLE:      The Univariate Marginal Distribution Algorithm Copes Well With Deception and Epistasis
http://arxiv.org/abs/2007.08277
AUTHORS:        Benjamin Doerr ; Martin S. Krejca
HIGHLIGHT:      In this work, we show that this negative finding is caused by an unfortunate choice of the parameters of the UMDA.

16, TITLE:      Radiation pattern prediction for Metasurfaces: A Neural Network based approach
http://arxiv.org/abs/2007.08035
AUTHORS:        Hamidreza Taghvaee ; Akshay Jain ; Xavier Timoneda ; Christos Liaskos ; Sergi Abadal ; Eduard Alarcón ; Albert Cabellos-Aparicio
COMMENTS:       Submitted to IEEE OJ-COMS
HIGHLIGHT:      Hence, in this paper we propose a novel neural networks based approach that enables a fast and accurate characterization of the MSF response.

17, TITLE:      On the Capability of Neural Networks to Generalize to Unseen Category-Pose Combinations
http://arxiv.org/abs/2007.08032
AUTHORS:        Spandan Madan ; Timothy Henry ; Jamell Dozier ; Helen Ho ; Nishchal Bhandari ; Tomotake Sasaki ; Frédo Durand ; Hanspeter Pfister ; Xavier Boix
HIGHLIGHT:      In this paper, we answer these questions by analyzing state-of-the-art DNNs trained to recognize both object category and pose (position, scale, and 3D viewpoint) with quantitative control over the number of category-pose combinations seen during training.

18, TITLE:      Kernelized Memory Network for Video Object Segmentation
http://arxiv.org/abs/2007.08270
AUTHORS:        Hongje Seong ; Junhyuk Hyun ; Euntai Kim
COMMENTS:       Accepted to ECCV 2020
HIGHLIGHT:      To solve the mismatch between STM and VOS, we propose a kernelized memory network (KMN).

19, TITLE:      Do Adversarially Robust ImageNet Models Transfer Better?
http://arxiv.org/abs/2007.08489
AUTHORS:        Hadi Salman ; Andrew Ilyas ; Logan Engstrom ; Ashish Kapoor ; Aleksander Madry
HIGHLIGHT:      In this work, we identify another such aspect: we find that adversarially robust models, while less accurate, often perform better than their standard-trained counterparts when used for transfer learning.

20, TITLE:      Learning to Restore a Single Face Image Degraded by Atmospheric Turbulence using CNNs
http://arxiv.org/abs/2007.08404
AUTHORS:        Rajeev Yasarla ; Vishal M Patel
HIGHLIGHT:      We present a deep learning-based solution to the problem of restoring a turbulence-degraded face image where prior information regarding the amount of geometric distortion and blur at each location of the face image is first estimated using two separate networks.

21, TITLE:      Autoregressive Unsupervised Image Segmentation
http://arxiv.org/abs/2007.08247
AUTHORS:        Yassine Ouali ; Céline Hudelot ; Myriam Tami
COMMENTS:       Accepted at ECCV 2020
HIGHLIGHT:      In this work, we propose a new unsupervised image segmentation approach based on mutual information maximization between different constructed views of the inputs.

22, TITLE:      Complete & Label: A Domain Adaptation Approach to Semantic Segmentation of LiDAR Point Clouds
http://arxiv.org/abs/2007.08488
AUTHORS:        Li Yi ; Boqing Gong ; Thomas Funkhouser
HIGHLIGHT:      We study an unsupervised domain adaptation problem for the semantic labeling of 3D point clouds, with a particular focus on domain discrepancies induced by different LiDAR sensors.

23, TITLE:      The role of collider bias in understanding statistics on racially biased policing
http://arxiv.org/abs/2007.08406
AUTHORS:        Norman Fenton ; Martin Neil ; Steven Frazier
COMMENTS:       7 pages, 5 figures
HIGHLIGHT:      We provide a causal Bayesian network model to explain this bias, which is called collider bias or Berkson's paradox, and show how the different conclusions arise from the same model and data.

24, TITLE:      D2D: Learning to find good correspondences for image matching and manipulation
http://arxiv.org/abs/2007.08480
AUTHORS:        Olivia Wiles ; Sebastien Ehrhardt ; Andrew Zisserman
HIGHLIGHT:      We propose a new approach to determining correspondences between image pairs under large changes in illumination, viewpoint, context, and material.

25, TITLE:      $λ_S$: Computable semantics for differentiable programming with higher-order functions and datatypes
http://arxiv.org/abs/2007.08017
AUTHORS:        Benjamin Sherman ; Jesse Michel ; Michael Carbin
COMMENTS:       41 pages, 10 figures
HIGHLIGHT:      We present a differentiable programming language, $\lambda_S$, that is the first to deliver a semantics for higher-order functions, higher-order derivatives, and Lipschitz but nondifferentiable functions.

26, TITLE:      Combining Task Predictors via Enhancing Joint Predictability
http://arxiv.org/abs/2007.08012
AUTHORS:        Kwang In Kim ; Christian Richardt ; Hyung Jin Chang
HIGHLIGHT:      We present a new predictor combination algorithm that improves the target by i) measuring the relevance of references based on their capabilities in predicting the target, and ii) strengthening such estimated relevance.

27, TITLE:      LogiQA: A Challenge Dataset for Machine Reading Comprehension with Logical Reasoning
http://arxiv.org/abs/2007.08124
AUTHORS:        Jian Liu ; Leyang Cui ; Hanmeng Liu ; Dandan Huang ; Yile Wang ; Yue Zhang
COMMENTS:       Accepted by IJCAI2020
HIGHLIGHT:      With the rising of deep learning techniques, algorithmic models rival human performances on simple QA, and thus increasingly challenging machine reading datasets have been proposed. We build a comprehensive dataset, named LogiQA, which is sourced from expert-written questions for testing human Logical reasoning.

28, TITLE:      Co-generation of game levels and game-playing agents
http://arxiv.org/abs/2007.08497
AUTHORS:        Aaron Dharna ; Julian Togelius ; L. B. Soros
COMMENTS:       7 pages, 5 figures, in-review
HIGHLIGHT:      This paper introduces a POET-Inspired Neuroevolutionary System for KreativitY (PINSKY) in games, which co-generates levels for multiple video games and agents that play them.

29, TITLE:      Vehicle Detection of Multi-source Remote Sensing Data Using Active Fine-tuning Network
http://arxiv.org/abs/2007.08494
AUTHORS:        Xin Wu ; Wei Li ; Danfeng Hong ; Jiaojiao Tian ; Ran Tao ; Qian Du
HIGHLIGHT:      Vehicle Detection of Multi-source Remote Sensing Data Using Active Fine-tuning Network

30, TITLE:      Efficient State Abstraction using Object-centered Predicates for Manipulation Planning
http://arxiv.org/abs/2007.08251
AUTHORS:        Alejandro Agostini ; Dongheui Lee
HIGHLIGHT:      To tackle these limitations we propose an object-centered representation that permits characterizing a much wider set of possible changes in configuration spaces than the traditional observer perspective counterpart.

31, TITLE:      Neuro-Endo-Trainer-Online Assessment System (NET-OAS) for Neuro-Endoscopic Skills Training
http://arxiv.org/abs/2007.08378
AUTHORS:        Vinkle Srivastav ; Britty Baby ; Ramandeep Singh ; Prem Kalra ; Ashish Suri
COMMENTS:       Published at Federated Conference on Computer Science and Information Systems - FedCSIS 2017
HIGHLIGHT:      The objective of the current study was to develop a modified version (Neuro-Endo-Trainer-Online Assessment System (NET-OAS)) by providing a stand-alone system with online evaluation and real-time feedback.

32, TITLE:      Interactive Video Object Segmentation Using Global and Local Transfer Modules
http://arxiv.org/abs/2007.08139
AUTHORS:        Yuk Heo ; Yeong Jun Koh ; Chang-Su Kim
HIGHLIGHT:      An interactive video object segmentation algorithm, which takes scribble annotations on query objects as input, is proposed in this paper.

33, TITLE:      Self-Supervised Nuclei Segmentation in Histopathological Images Using Attention
http://arxiv.org/abs/2007.08373
AUTHORS:        Mihir Sahasrabudhe ; Stergios Christodoulidis ; Roberto Salgado ; Stefan Michiels ; Sherene Loi ; Fabrice André ; Nikos Paragios ; Maria Vakalopoulou
COMMENTS:       10 pages. Code available online at https://github.com/msahasrabudhe/miccai2020_self_sup_nuclei_seg
HIGHLIGHT:      In this study, we present a self-supervised approach for segmentation of nuclei for whole slide histopathology images.

34, TITLE:      Layer-Wise Adaptive Updating for Few-Shot Image Classification
http://arxiv.org/abs/2007.08129
AUTHORS:        Yunxiao Qin ; Weiguo Zhang ; Zezheng Wang ; Chenxu Zhao ; Jingping Shi
HIGHLIGHT:      In this paper, we propose a novel meta-learning based layer-wise adaptive updating (LWAU) method for FSIC.

35, TITLE:      Probabilistic Anchor Assignment with IoU Prediction for Object Detection
http://arxiv.org/abs/2007.08103
AUTHORS:        Kang Kim ; Hee Seok Lee
COMMENTS:       ECCV 2020
HIGHLIGHT:      In this paper we propose a novel anchor assignment strategy that adaptively separates anchors into positive and negative samples for a ground truth bounding box according to the model's learning status such that it is able to reason about the separation in a probabilistic manner.

36, TITLE:      Distributed Reinforcement Learning of Targeted Grasping with Active Vision for Mobile Manipulators
http://arxiv.org/abs/2007.08082
AUTHORS:        Yasuhiro Fujita ; Kota Uenishi ; Avinash Ummadisingu ; Prabhat Nagarajan ; Shimpei Masuda ; Mario Ynocente Castro
COMMENTS:       Accepted at IROS 2020
HIGHLIGHT:      To achieve such a system, we combine several advances in deep reinforcement learning and present a large-scale distributed training system using synchronous SGD that seamlessly scales to multi-node, multi-GPU infrastructure to make rapid prototyping easier.

37, TITLE:      Towards Debiasing Sentence Representations
http://arxiv.org/abs/2007.08100
AUTHORS:        Paul Pu Liang ; Irene Mengze Li ; Emily Zheng ; Yao Chong Lim ; Ruslan Salakhutdinov ; Louis-Philippe Morency
COMMENTS:       ACL 2020, code available at https://github.com/pliang279/sent_debias
HIGHLIGHT:      In this paper, we investigate the presence of social biases in sentence-level representations and propose a new method, Sent-Debias, to reduce these biases.

38, TITLE:      Dueling Deep Q Network for Highway Decision Making in Autonomous Vehicles: A Case Study
http://arxiv.org/abs/2007.08343
AUTHORS:        Teng Liu ; Xingyu Mu ; Xiaolin Tang ; Bing Huang ; Hong Wang ; Dongpu Cao
COMMENTS:       5 pages, 6 figures
HIGHLIGHT:      Dueling Deep Q Network for Highway Decision Making in Autonomous Vehicles: A Case Study

39, TITLE:      TrashCan: A Semantically-Segmented Dataset towards Visual Detection of Marine Debris
http://arxiv.org/abs/2007.08097
AUTHORS:        Jungseok Hong ; Michael Fulton ; Junaed Sattar
HIGHLIGHT:      This paper presents TrashCan, a large dataset comprised of images of underwater trash collected from a variety of sources, annotated both using bounding boxes and segmentation labels, for development of robust detectors of marine debris.

40, TITLE:      Human Pose Estimation on Privacy-Preserving Low-Resolution Depth Images
http://arxiv.org/abs/2007.08340
AUTHORS:        Vinkle Srivastav ; Afshin Gangi ; Nicolas Padoy
COMMENTS:       Published at MICCAI-2019
HIGHLIGHT:      In this paper, we introduce the problem of HPE on low-resolution depth images and propose an end-to-end solution that integrates a multi-scale super-resolution network with a 2D human pose estimation network.

41, TITLE:      PerMO: Perceiving More at Once from a Single Image for Autonomous Driving
http://arxiv.org/abs/2007.08116
AUTHORS:        Feixiang Lu ; Zongdai Liu ; Xibin Song ; Dingfu Zhou ; Wei Li ; Hui Miao ; Miao Liao ; Liangjun Zhang ; Bin Zhou ; Ruigang Yang ; Dinesh Manocha
HIGHLIGHT:      We present a novel approach to detect, segment, and reconstruct complete textured 3D models of vehicles from a single image for autonomous driving.

42, TITLE:      Defocus Blur Detection via Depth Distillation
http://arxiv.org/abs/2007.08113
AUTHORS:        Xiaodong Cun ; Chi-Man Pun
COMMENTS:       ECCV 2020
HIGHLIGHT:      Hence, we consider the depth information as the approximate soft label of DBD and propose a joint learning framework inspired by knowledge distillation.

43, TITLE:      Self-supervision on Unlabelled OR Data for Multi-person 2D/3D Human Pose Estimation
http://arxiv.org/abs/2007.08354
AUTHORS:        Vinkle Srivastav ; Afshin Gangi ; Nicolas Padoy
COMMENTS:       Accepted at MICCAI 2020
HIGHLIGHT:      In this work, we propose to use knowledge distillation in a teacher/student framework to harness the knowledge present in a large-scale non-annotated dataset and in an accurate but complex multi-stage teacher network to train a lightweight network for joint 2D/3D pose estimation.

44, TITLE:      Strengthening Deterministic Policies for POMDPs
http://arxiv.org/abs/2007.08351
AUTHORS:        Leonore Winterer ; Ralf Wimmer ; Nils Jansen ; Bernd Becker
HIGHLIGHT:      We provide a novel MILP encoding that supports sophisticated specifications in the form of temporal logic constraints.

45, TITLE:      Natural Graph Networks
http://arxiv.org/abs/2007.08349
AUTHORS:        Pim de Haan ; Taco Cohen ; Max Welling
HIGHLIGHT:      Studying the local symmetries of graphs, we propose a more general algorithm that uses different kernels on different edges, making the network equivariant to local and global graph isomorphisms and hence more expressive.

46, TITLE:      Fighting the COVID-19 Infodemic in Social Media: A Holistic Perspective and a Call to Arms
http://arxiv.org/abs/2007.07996
AUTHORS:        Firoj Alam ; Fahim Dalvi ; Shaden Shaar ; Nadir Durrani ; Hamdy Mubarak ; Alex Nikolov ; Giovanni Da San Martino ; Ahmed Abdelali ; Hassan Sajjad ; Kareem Darwish ; Preslav Nakov
COMMENTS:       COVID-19, Infodemic, Disinformation, Misinformation, Fake News, Call to Arms, Crowdsourcing Annotations
HIGHLIGHT:      Now, we issue a call to arms to the research community and beyond to join the fight by supporting our crowdsourcing annotation efforts.

47, TITLE:      Boosting Weakly Supervised Object Detection with Progressive Knowledge Transfer
http://arxiv.org/abs/2007.07986
AUTHORS:        Yuanyi Zhong ; Jianfeng Wang ; Jian Peng ; Lei Zhang
COMMENTS:       ECCV 2020. Code: https://github.com/mikuhatsune/wsod_transfer
HIGHLIGHT:      In this paper, we propose an effective knowledge transfer framework to boost the weakly supervised object detection accuracy with the help of an external fully-annotated source dataset, whose categories may not overlap with the target domain.

48, TITLE:      Overview of CheckThat! 2020: Automatic Identification and Verification of Claims in Social Media
http://arxiv.org/abs/2007.07997
AUTHORS:        Alberto Barron-Cedeno ; Tamer Elsayed ; Preslav Nakov ; Giovanni Da San Martino ; Maram Hasanain ; Reem Suwaileh ; Fatima Haouari ; Nikolay Babulkov ; Bayan Hamdan ; Alex Nikolov ; Shaden Shaar ; Zien Sheikh Ali
COMMENTS:       Check-Worthiness Estimation, Fact-Checking, Veracity, Evidence-based Verification, Detecting Previously Fact-Checked Claims, Social Media Verification, Computational Journalism, COVID-19
HIGHLIGHT:      We present an overview of the third edition of the CheckThat! Last but not least, we release to the research community all datasets from the lab as well as the evaluation scripts, which should enable further research in the important tasks of check-worthiness estimation and automatic claim verification.

49, TITLE:      Compression strategies and space-conscious representations for deep neural networks
http://arxiv.org/abs/2007.07967
AUTHORS:        Giosuè Cataldo Marinò ; Gregorio Ghidoli ; Marco Frasca ; Dario Malchiodi
HIGHLIGHT:      In this paper, we investigate the impact of lossy compression of CNNs by weight pruning and quantization, and lossless weight matrix representations based on source coding.

50, TITLE:      SAILenv: Learning in Virtual Visual Environments Made Simple
http://arxiv.org/abs/2007.08224
AUTHORS:        Enrico Meloni ; Luca Pasqualini ; Matteo Tiezzi ; Marco Gori ; Stefano Melacci
COMMENTS:       8 pages, 7 figures, submitted to ICPR 2020
HIGHLIGHT:      In this paper, we present a novel platform, SAILenv, that is specifically designed to be simple and customizable, and that allows researchers to experiment visual recognition in virtual 3D scenes.

51, TITLE:      Separating Sounds from a Single Image
http://arxiv.org/abs/2007.07984
AUTHORS:        Lingyu Zhu ; Esa Rahtu
COMMENTS:       main paper 6 pages, ref 1 page. Project page: https://ly-zhu.github.io/separating-sounds-from-single-image
HIGHLIGHT:      In this paper, we investigate the performance of appearance information, extracted from a single image, in the task of recovering the original component signals from a mixture audio.

52, TITLE:      openDD: A Large-Scale Roundabout Drone Dataset
http://arxiv.org/abs/2007.08463
AUTHORS:        Antonia Breuer ; Jan-Aike Termöhlen ; Silviu Homoceanu ; Tim Fingscheidt
COMMENTS:       ITSC 2020 Conference Paper
HIGHLIGHT:      We introduce the openDD dataset, including 84,774 accurately tracked trajectories and HD map data of seven different roundabouts.

53, TITLE:      DRIFT: Deep Reinforcement Learning for Functional Software Testing
http://arxiv.org/abs/2007.08220
AUTHORS:        Luke Harries ; Rebekah Storan Clarke ; Timothy Chapman ; Swamy V. P. L. N. Nallamalli ; Levent Ozgur ; Shuktika Jain ; Alex Leung ; Steve Lim ; Aaron Dietrich ; José Miguel Hernández-Lobato ; Tom Ellis ; Cheng Zhang ; Kamil Ciosek
HIGHLIGHT:      In this work, we propose a Reinforcement Learning (RL) framework for functional software testing named DRIFT.

54, TITLE:      An Efficient Mixture of Deep and Machine Learning Models for COVID-19 and Tuberculosis Detection Using X-Ray Images in Resource Limited Settings
http://arxiv.org/abs/2007.08223
AUTHORS:        Ali H. Al-Timemy ; Rami N. Khushaba ; Zahraa M. Mosa ; Javier Escudero
COMMENTS:       The final constructed dataset named COVID-19 five-class balanced dataset is available from: https://drive.google.com/drive/folders/1toMymyHTy0DR_fyE7hjO3LSBGWtVoPNf?usp=sharing
HIGHLIGHT:      In order to help in the detection of COVID-19, we propose the extraction of deep features (DF) from chest X-ray images, a technology available in most hospitals, and their subsequent classification using machine learning methods that do not require large computational resources.

55, TITLE:      Inheritance software metrics on smart contracts
http://arxiv.org/abs/2007.08222
AUTHORS:        Ashish Rajendra Sai ; Conor Holmes ; Jim Buckley ; Andrew Le Gear
COMMENTS:       Accepted by International Conference on Program Comprehension (ICPC 2020)
HIGHLIGHT:      In this paper, we empirically evaluate inheritance-based metrics as applied to smart contracts.

56, TITLE:      CloudCast: A Satellite-Based Dataset and Baseline for Forecasting Clouds
http://arxiv.org/abs/2007.07978
AUTHORS:        A. H. Nielsen ; A. Iosifidis ; H. Karstoft
COMMENTS:       For the novel dataset, see https://vision.eng.au.dk/cloudcast-dataset/
HIGHLIGHT:      In this paper, we present a novel satellite-based dataset called "CloudCast".

57, TITLE:      PC-PG: Policy Cover Directed Exploration for Provable Policy Gradient Learning
http://arxiv.org/abs/2007.08459
AUTHORS:        Alekh Agarwal ; Mikael Henaff ; Sham Kakade ; Wen Sun
HIGHLIGHT:      This work introduces the the Policy Cover-Policy Gradient (PC-PG) algorithm, which provably balances the exploration vs. exploitation tradeoff using an ensemble of learned policies (the policy cover).

58, TITLE:      The Notary in the Haystack -- Countering Class Imbalance in Document Processing with CNNs
http://arxiv.org/abs/2007.07943
AUTHORS:        Martin Leipert ; Georg Vogeler ; Mathias Seuret ; Andreas Maier ; Vincent Christlein
COMMENTS:       Accepted at DAS Workshop 2020
HIGHLIGHT:      In this work, we evaluate different countermeasures for this problem.

59, TITLE:      3D CNN-PCA: A Deep-Learning-Based Parameterization for Complex Geomodels
http://arxiv.org/abs/2007.08478
AUTHORS:        Yimin Liu ; Louis J. Durlofsky
HIGHLIGHT:      In this study, a deep-learning-based geological parameterization algorithm, CNN-PCA, is developed for complex 3D geomodels.

60, TITLE:      U-Net Based Architecture for an Improved Multiresolution Segmentation in Medical Images
http://arxiv.org/abs/2007.08238
AUTHORS:        Simindokht Jahangard ; Mohammad Hossein Zangoei ; Maysam Shahedi
HIGHLIGHT:      In this study, our objective is to improve the multi-resolution image segmentation performance of U-Net architecture.

61, TITLE:      EfficientHRNet: Efficient Scaling for Lightweight High-Resolution Multi-Person Pose Estimation
http://arxiv.org/abs/2007.08090
AUTHORS:        Christopher Neff ; Aneri Sheth ; Steven Furgurson ; Hamed Tabkhi
COMMENTS:       14 pages (18 with references), 3 figures
HIGHLIGHT:      In this paper, we present EfficientHRNet, a family of lightweight 2D human pose estimators that unifies the high-resolution structure of state-of-the-art HigherHRNet with the highly efficient model scaling principles of EfficientNet to create high accuracy models with significantly reduced computation costs compared to other state-of-the-art approaches.

62, TITLE:      Provable Worst Case Guarantees for the Detection of Out-of-Distribution Data
http://arxiv.org/abs/2007.08473
AUTHORS:        Julian Bitterwolf ; Alexander Meinke ; Matthias Hein
COMMENTS:       Code available at https://gitlab.com/Bitterwolf/GOOD
HIGHLIGHT:      In this paper, we are aiming for certifiable worst case guarantees for OOD detection by enforcing not only low confidence at the OOD point but also in an $l_\infty$-ball around it.

63, TITLE:      Maximizing coverage while ensuring fairness: a tale of conflicting objective
http://arxiv.org/abs/2007.08069
AUTHORS:        Abolfazl Asudeh ; Tanya Berger-Wolf ; Bhaskar DasGupta ; Anastasios Sidiropoulos
HIGHLIGHT:      In this paper we address the problem of incorporation of fairness from a $combinatorial$ $optimization$ perspective.

64, TITLE:      Mixture of Step Returns in Bootstrapped DQN
http://arxiv.org/abs/2007.08229
AUTHORS:        Po-Han Chiang ; Hsuan-Kung Yang ; Zhang-Wei Hong ; Chun-Yi Lee
HIGHLIGHT:      To address this issue, we propose Mixture Bootstrapped DQN (MB-DQN) built on top of bootstrapped DQN, and uses different backup lengths for different bootstrapped heads. We further provide a set of ablation studies to examine the impacts of different design configurations of MB-DQN.

65, TITLE:      Hierarchical Interaction Networks with Rethinking Mechanism for Document-level Sentiment Analysis
http://arxiv.org/abs/2007.08445
AUTHORS:        Lingwei Wei ; Dou Hu ; Wei Zhou ; Xuehai Tang ; Xiaodan Zhang ; Xin Wang ; Jizhong Han ; Songlin Hu
COMMENTS:       16 pages, accepted by ECML-PKDD 2020
HIGHLIGHT:      In this paper, we study how to effectively generate a discriminative representation with explicit subject patterns and sentiment contexts for DSA.

66, TITLE:      Provably Good Batch Reinforcement Learning Without Great Exploration
http://arxiv.org/abs/2007.08202
AUTHORS:        Yao Liu ; Adith Swaminathan ; Alekh Agarwal ; Emma Brunskill
COMMENTS:       36 pages, 7 figures
HIGHLIGHT:      Batch reinforcement learning (RL) is important to apply RL algorithms to many high stakes tasks.

67, TITLE:      Suppress and Balance: A Simple Gated Network for Salient Object Detection
http://arxiv.org/abs/2007.08074
AUTHORS:        Xiaoqi Zhao ; Youwei Pang ; Lihe Zhang ; Huchuan Lu ; Lei Zhang
COMMENTS:       Accepted in ECCV2020(oral). Code: https://github.com/Xiaoqi-Zhao-DLUT/GateNet-RGB-Saliency
HIGHLIGHT:      In this work, we propose a simple gated network (GateNet) to solve both issues at once.

68, TITLE:      Unseen Object Instance Segmentation for Robotic Environments
http://arxiv.org/abs/2007.08073
AUTHORS:        Christopher Xie ; Yu Xiang ; Arsalan Mousavian ; Dieter Fox
COMMENTS:       Extended version of arXiv:1907.13236
HIGHLIGHT:      To train our method, we introduce a large-scale synthetic dataset of random objects on tabletops.

69, TITLE:      Memory Based Attentive Fusion
http://arxiv.org/abs/2007.08076
AUTHORS:        Darshana Priyasad ; Tharindu Fernando ; Simon Denman ; Sridha Sridharan ; Clinton Fookes
COMMENTS:       Pre-print submitted to Information Fusion
HIGHLIGHT:      In this paper, we present a novel Memory Based Attentive Fusion (MBAF) layer, which fuses modes by incorporating both the current features and long-term dependencies in the data, thus allowing the model to understand the relative importance of modes over time.

70, TITLE:      Kronecker Attention Networks
http://arxiv.org/abs/2007.08442
AUTHORS:        Hongyang Gao ; Zhengyang Wang ; Shuiwang Ji
COMMENTS:       9 pages, KDD2020
HIGHLIGHT:      In this work, we propose to avoid flattening by assuming the data follow matrix-variate normal distributions.

71, TITLE:      Learning End-to-End Action Interaction by Paired-Embedding Data Augmentation
http://arxiv.org/abs/2007.08071
AUTHORS:        Ziyang Song ; Zejian Yuan ; Chong Zhang ; Wanchao Chi ; Yonggen Ling ; Shenghao Zhang
COMMENTS:       16 pages, 7 figures
HIGHLIGHT:      In this paper, we specify a new Interactive Action Translation (IAT) task which aims to learn end-to-end action interaction from unlabeled interactive pairs, removing explicit action recognition.

72, TITLE:      SafeRESTScript: Statically Checking REST API Consumers
http://arxiv.org/abs/2007.08048
AUTHORS:        Nuno Burnay ; Antónia Lopes ; Vasco T. Vasconcelos
HIGHLIGHT:      In this paper, we present SafeRESTScript (SRS, for short) a language that extends the support of static analysis to calls to REST services, with the ability to statically find common errors such as missing or invalid data in REST calls and misuse of the results from such calls.

73, TITLE:      Negative Pseudo Labeling using Class Proportion for Semantic Segmentation in Pathology
http://arxiv.org/abs/2007.08044
AUTHORS:        Hiroki Tokunaga ; Brian Kenji Iwana ; Yuki Teramoto ; Akihiko Yoshizawa ; Ryoma Bise
COMMENTS:       17 pages, 7 figures, Accepted in ECCV 2020
HIGHLIGHT:      We propose a weakly-supervised cell tracking method that can train a convolutional neural network (CNN) by using only the annotation of "cell detection" (i.e., the coordinates of cell positions) without association information, in which cell positions can be easily obtained by nuclear staining.

74, TITLE:      DeepInit Phase Retrieval
http://arxiv.org/abs/2007.08214
AUTHORS:        Martin Reiche ; Peter Jung
COMMENTS:       9 pages, 12 figures
HIGHLIGHT:      We therefore propose DeepInit Phase Retrieval, which uses regularized gradient descent under a deep generative data prior to compute a trained initialization for a fast classical algorithm (e.g. the randomized Kaczmarz method).

75, TITLE:      Video-based Remote Physiological Measurement via Cross-verified Feature Disentangling
http://arxiv.org/abs/2007.08213
AUTHORS:        Xuesong Niu ; Zitong Yu ; Hu Han ; Xiaobai Li ; Shiguang Shan ; Guoying Zhao
HIGHLIGHT:      Video-based Remote Physiological Measurement via Cross-verified Feature Disentangling

76, TITLE:      Black-Box Watermarking for Generative Adversarial Networks
http://arxiv.org/abs/2007.08457
AUTHORS:        Vladislav Skripniuk ; Ning Yu ; Sahar Abdelnabi ; Mario Fritz
HIGHLIGHT:      We propose the first watermarking solution for GAN models.

77, TITLE:      Event Enhanced High-Quality Image Recovery
http://arxiv.org/abs/2007.08336
AUTHORS:        Bishan Wang ; Jingwei He ; Lei Yu ; Gui-Song Xia ; Wen Yang
HIGHLIGHT:      Based on this, we propose an explainable network, an event-enhanced sparse learning network (eSL-Net), to recover the high-quality images from event cameras.

78, TITLE:      Specification mining and automated task planning for autonomous robots based on a graph-based spatial temporal logic
http://arxiv.org/abs/2007.08451
AUTHORS:        Zhiyu Liu ; Meng Jiang ; Hai Lin
HIGHLIGHT:      We aim to enable an autonomous robot to learn new skills from demo videos and use these newly learned skills to accomplish non-trivial high-level tasks.

79, TITLE:      Shape Prior Deformation for Categorical 6D Object Pose and Size Estimation
http://arxiv.org/abs/2007.08454
AUTHORS:        Meng Tian ; Marcelo H Ang Jr ; Gim Hee Lee
COMMENTS:       Accepted at ECCV 2020
HIGHLIGHT:      We present a novel learning approach to recover the 6D poses and sizes of unseen object instances from an RGB-D image.

80, TITLE:      Towards Evaluating Driver Fatigue with Robust Deep Learning Models
http://arxiv.org/abs/2007.08453
AUTHORS:        Ken Alparslan ; Yigit Alparslan ; Matthew Burlick
COMMENTS:       8 pages, 12 figures
HIGHLIGHT:      In this paper, we explore different deep learning based approaches to detect driver fatigue.

81, TITLE:      Translate Reverberated Speech to Anechoic Ones: Speech Dereverberation with BERT
http://arxiv.org/abs/2007.08052
AUTHORS:        Yang Jiao
HIGHLIGHT:      Single channel speech dereverberation is considered in this work.

82, TITLE:      A Genetic Algorithm for Obtaining Memory Constrained Near-Perfect Hashing
http://arxiv.org/abs/2007.08311
AUTHORS:        Dan Domnita ; Ciprian Oprisa
HIGHLIGHT:      We present an approach based on hash tables that focuses on both minimizing the number of comparisons performed during the search and minimizing the total collection size.

83, TITLE:      Predicting Mechanical Ventilation Requirement and Mortality in COVID-19 using Radiomics and Deep Learning on Chest Radiographs: A Multi-Institutional Study
http://arxiv.org/abs/2007.08028
AUTHORS:        Joseph Bae ; Saarthak Kapse ; Gagandeep Singh ; Tej Phatak ; Jeremy Green ; Nikhil Madan ; Prateek Prasanna
COMMENTS:       Joseph Bae and Saarthak Kapse have contributed equally to this work. This manuscript has been submitted for consideration to _European Radiology_. 12 pages, 6 figures, 4 tables
HIGHLIGHT:      The models proposed in this study and the prognostic information they provide, complementary to other clinical data, might be used to aid physician decision making and resource allocation during the COVID-19 pandemic.

84, TITLE:      A Survey on Computational Propaganda Detection
http://arxiv.org/abs/2007.08024
AUTHORS:        Giovanni Da San Martino ; Stefano Cresci ; Alberto Barron-Cedeno ; Seunghak Yu ; Roberto Di Pietro ; Preslav Nakov
COMMENTS:       propaganda detection, disinformation, misinformation, fake news, media bias
HIGHLIGHT:      In this survey, we review the state of the art on computational propaganda detection from the perspective of Natural Language Processing and Network Analysis, arguing about the need for combined efforts between these communities.

85, TITLE:      Weighing Counts: Sequential Crowd Counting by Reinforcement Learning
http://arxiv.org/abs/2007.08260
AUTHORS:        Liang Liu ; Hao Lu ; Hongwei Zou ; Haipeng Xiong ; Zhiguo Cao ; Chunhua Shen
COMMENTS:       Accepted to Proc. Eur. Conf. Computer Vision (ECCV) 2020
HIGHLIGHT:      Inspired by scale weighing, we propose a novel 'counting scale' termed LibraNet where the count value is analogized by weight.

86, TITLE:      Efficient Full Image Interactive Segmentation by Leveraging Within-image Appearance Similarity
http://arxiv.org/abs/2007.08173
AUTHORS:        Mykhaylo Andriluka ; Stefano Pellegrini ; Stefan Popov ; Vittorio Ferrari
HIGHLIGHT:      We propose a new approach to interactive full-image semantic segmentation which enables quickly collecting training data for new datasets with previously unseen semantic classes (A demo is available at https://youtu.be/yUk8D5gEX-o).

87, TITLE:      VIPriors Object Detection Challenge
http://arxiv.org/abs/2007.08170
AUTHORS:        Zhipeng Luo ; Lixuan Che
HIGHLIGHT:      In this paper, we study analysis the characteristics of the data, and an effective data enhancement method is proposed.

88, TITLE:      Accelerating 3D Deep Learning with PyTorch3D
http://arxiv.org/abs/2007.08501
AUTHORS:        Nikhila Ravi ; Jeremy Reizenstein ; David Novotny ; Taylor Gordon ; Wan-Yen Lo ; Justin Johnson ; Georgia Gkioxari
COMMENTS:       tech report
HIGHLIGHT:      We address these challenges by introducing PyTorch3D, a library of modular, efficient, and differentiable operators for 3D deep learning.

89, TITLE:      Multi-Task Pruning for Semantic Segmentation Networks
http://arxiv.org/abs/2007.08386
AUTHORS:        Xinghao Chen ; Yunhe Wang ; Yiman Zhang ; Peng Du ; Chunjing Xu ; Chang Xu
HIGHLIGHT:      Hence to identify the redundancy in segmentation networks, we present a multi-task channel pruning approach.

90, TITLE:      RetrieveGAN: Image Synthesis via Differentiable Patch Retrieval
http://arxiv.org/abs/2007.08513
AUTHORS:        Hung-Yu Tseng ; Hsin-Ying Lee ; Lu Jiang ; Ming-Hsuan Yang ; Weilong Yang
COMMENTS:       ECCV 2020
HIGHLIGHT:      In this work, we aim to synthesize images from scene description with retrieved patches as reference.

91, TITLE:      Enhanced detection of fetal pose in 3D MRI by Deep Reinforcement Learning with physical structure priors on anatomy
http://arxiv.org/abs/2007.08146
AUTHORS:        Molin Zhang ; Junshen Xu ; Esra Abaci Turk ; P. Ellen Grant ; Polina Golland ; Elfar Adalsteinsson
COMMENTS:       10 pages, 3 figures, MICCAI 2020
HIGHLIGHT:      The optimization is challenging, and here we propose an improved DRL that incorporates priors on physical structure of the fetal body.

92, TITLE:      RepPoints V2: Verification Meets Regression for Object Detection
http://arxiv.org/abs/2007.08508
AUTHORS:        Yihong Chen ; Zheng Zhang ; Yue Cao ; Liwei Wang ; Stephen Lin ; Han Hu
HIGHLIGHT:      In this paper, we take this philosophy to improve state-of-the-art object detection, specifically by RepPoints.

93, TITLE:      World-Consistent Video-to-Video Synthesis
http://arxiv.org/abs/2007.08509
AUTHORS:        Arun Mallya ; Ting-Chun Wang ; Karan Sapra ; Ming-Yu Liu
COMMENTS:       Published at the European Conference on Computer Vision, 2020
HIGHLIGHT:      To address the limitation, we introduce a novel vid2vid framework that efficiently and effectively utilizes all past generated frames during rendering.

94, TITLE:      Odyssey: Creation, Analysis and Detection of Trojan Models
http://arxiv.org/abs/2007.08142
AUTHORS:        Marzieh Edraki ; Nazmul Karim ; Nazanin Rahnavard ; Ajmal Mian ; Mubarak Shah
HIGHLIGHT:      In this paper, we develop a detector based upon the analysis of intrinsic properties of DNN that could get affected by a Trojan attack.

95, TITLE:      Implicit Mesh Reconstruction from Unannotated Image Collections
http://arxiv.org/abs/2007.08504
AUTHORS:        Shubham Tulsiani ; Nilesh Kulkarni ; Abhinav Gupta
COMMENTS:       Project page: https://shubhtuls.github.io/imr/
HIGHLIGHT:      We present an approach to infer the 3D shape, texture, and camera pose for an object from a single RGB image, using only category-level image collections with foreground masks as supervision.

96, TITLE:      FeatMatch: Feature-Based Augmentation for Semi-Supervised Learning
http://arxiv.org/abs/2007.08505
AUTHORS:        Chia-Wen Kuo ; Chih-Yao Ma ; Jia-Bin Huang ; Zsolt Kira
COMMENTS:       Paper accepted in ECCV 2020. Project page: https://sites.google.com/view/chiawen-kuo/home/featmatch
HIGHLIGHT:      In this paper, we propose a novel learned feature-based refinement and augmentation method that produces a varied set of complex transformations.

97, TITLE:      Controllable Image Synthesis via SegVAE
http://arxiv.org/abs/2007.08397
AUTHORS:        Yen-Chi Cheng ; Hsin-Ying Lee ; Min Sun ; Ming-Hsuan Yang
COMMENTS:       ECCV 2020. Project page: https://yccyenchicheng.github.io/SegVAE/ Code: https://github.com/yccyenchicheng/SegVAE
HIGHLIGHT:      In this work, we specifically target at generating semantic maps given a label-set consisting of desired categories.

98, TITLE:      Comprehensive Facial Expression Synthesis using Human-Interpretable Language
http://arxiv.org/abs/2007.08154
AUTHORS:        Joanna Hong ; Jung Uk Kim ; Sangmin Lee ; Yong Man Ro
COMMENTS:       ICIP 2020
HIGHLIGHT:      In this paper, therefore, we propose a new facial expression synthesis model from language-based facial expression description.

99, TITLE:      Semi-Siamese Training for Shallow Face Learning
http://arxiv.org/abs/2007.08398
AUTHORS:        Hang Du ; Hailin Shi ; Yuchi Liu ; Jun Wang ; Zhen Lei ; Dan Zeng ; Tao Mei
COMMENTS:       ECCV 2020 Spotlight
HIGHLIGHT:      In this paper, we aim to address the problem by introducing a novel training method named Semi-Siamese Training (SST).

100, TITLE:     Coupling Distant Annotation and Adversarial Training for Cross-Domain Chinese Word Segmentation
http://arxiv.org/abs/2007.08186
AUTHORS:        Ning Ding ; Dingkun Long ; Guangwei Xu ; Muhua Zhu ; Pengjun Xie ; Xiaobin Wang ; Hai-Tao Zheng
COMMENTS:       Accepted by ACL 2020
HIGHLIGHT:      In order to simultaneously alleviate these two issues, this paper proposes to couple distant annotation and adversarial training for cross-domain CWS.

101, TITLE:     Challenge report:VIPriors Action Recognition Challenge
http://arxiv.org/abs/2007.08180
AUTHORS:        Zhipeng Luo ; Dawei Xu ; Zhiguang Zhang
COMMENTS:       ECCV2020,VIPriors Action Recognition Challenge
HIGHLIGHT:      In this paper, we study previous methods and propose our method.

102, TITLE:     Learning from Noisy Labels with Deep Neural Networks: A Survey
http://arxiv.org/abs/2007.08199
AUTHORS:        Hwanjun Song ; Minseok Kim ; Dongmin Park ; Jae-Gil Lee
COMMENTS:       This work has been submitted to the IEEE for possible publication. Copyright may be transferred without notice, after which this version may no longer be accessible
HIGHLIGHT:      Subsequently, we summarize the typically used evaluation methodology, including public noisy datasets and evaluation metrics.

103, TITLE:     Training Interpretable Convolutional Neural Networks by Differentiating Class-specific Filters
http://arxiv.org/abs/2007.08194
AUTHORS:        Haoyu Liang ; Zhihao Ouyang ; Yuyuan Zeng ; Hang Su ; Zihao He ; Shu-Tao Xia ; Jun Zhu ; Bo Zhang
COMMENTS:       14pages (except appendix and reference), 9 figures, 3 tables
HIGHLIGHT:      Inspired by cellular differentiation, we propose a novel strategy to train interpretable CNNs by encouraging class-specific filters, among which each filter responds to only one (or few) class.

104, TITLE:     A high fidelity synthetic face framework for computer vision
http://arxiv.org/abs/2007.08364
AUTHORS:        Tadas Baltrusaitis ; Erroll Wood ; Virginia Estellers ; Charlie Hewitt ; Sebastian Dziadzio ; Marek Kowalski ; Matthew Johnson ; Thomas J. Cashman ; Jamie Shotton
HIGHLIGHT:      In our work we propose synthesizing such facial data, including ground truth annotations that would be almost impossible to acquire through manual annotation at the consistency and scale possible through use of synthetic data.



==========Updates to Previous Papers==========
1, TITLE:       Learning to Exploit Multiple Vision Modalities by Using Grafted Networks
http://arxiv.org/abs/2003.10959
AUTHORS:        Yuhuang Hu ; Tobi Delbruck ; Shih-Chii Liu
COMMENTS:       Accepted at ECCV 2020, 14 pages
HIGHLIGHT:      This paper proposes a Network Grafting Algorithm (NGA), where a new front end network driven by unconventional visual inputs replaces the front end network of a pretrained deep network that processes intensity frames.

2, TITLE:       Evolving Robust Neural Architectures to Defend from Adversarial Attacks
http://arxiv.org/abs/1906.11667
AUTHORS:        Shashank Kotyan ; Danilo Vasconcellos Vargas
COMMENTS:       Pre-print of the published article in Proceedings of the Workshop on Artificial Intelligence Safety 2020, co-located with the 29th International Joint Conference on Artificial Intelligence and the 17th Pacific Rim International Conference on Artificial Intelligence (IJCAI-PRICAI 2020)
HIGHLIGHT:      Here, we propose to use adversarial attacks as a function evaluation to search for neural architectures that can resist such attacks automatically.

3, TITLE:       Complexity of Maximum Cut on Interval Graphs
http://arxiv.org/abs/2006.00061
AUTHORS:        Ranendu Adhikary ; Kaustav Bose ; Satwik Mukherjee ; Bodhayan Roy
HIGHLIGHT:      We resolve the longstanding open problem concerning the computational complexity of Max Cut on interval graphs by showing that it is NP-complete.

4, TITLE:       SS-CAM: Smoothed Score-CAM for Sharper Visual Feature Localization
http://arxiv.org/abs/2006.14255
AUTHORS:        Haofan Wang ; Rakshit Naidu ; Joy Michael ; Soumya Snigdha Kundu
COMMENTS:       7 pages, 4 figures and 4 tables
HIGHLIGHT:      In this paper, built on the top of Score-CAM, we introduce an enhanced visual explanation in terms of visual sharpness called SS-CAM, which produces centralized localization of object features within an image through a smooth operation.

5, TITLE:       Face Identification using Local Ternary Tree Pattern based Spatial Structural Components
http://arxiv.org/abs/1905.00693
AUTHORS:        Rinku Datta Rakshit ; Dakshina Ranjan Kisku ; Massimo Tistarelli ; Phalguni Gupta
COMMENTS:       13 pages, 5 figures, conference paper
HIGHLIGHT:      This paper reports a face identification system which makes use of a novel local descriptor called Local Ternary Tree Pattern (LTTP).

6, TITLE:       Applying Deep-Learning-Based Computer Vision to Wireless Communications: Methodologies, Opportunities, and Challenges
http://arxiv.org/abs/2006.05782
AUTHORS:        Yu Tian ; Gaofeng Pan ; Mohamed-Slim Alouini
HIGHLIGHT:      The primary purpose of this article, then, is to introduce ideas about applying DL-based CV in wireless communications to bring some novel degrees of freedom to both theoretical research and engineering applications.

7, TITLE:       Joint Cross-Modality Super Resolution
http://arxiv.org/abs/2004.09965
AUTHORS:        Guy Shacht ; Sharon Fogel ; Dov Danon ; Daniel Cohen-Or ; Ilya Leizerson
HIGHLIGHT:      To this end, Cross-Modality Super-Resolution methods were introduced, where an RGB image of a high-resolution assists in increasing the resolution of the low-resolution modality.

8, TITLE:       Human-Expert-Level Brain Tumor Detection Using Deep Learning with Data Distillation and Augmentation
http://arxiv.org/abs/2006.12285
AUTHORS:        Diyuan Lu ; Nenad Polomac ; Iskra Gacheva ; Elke Hattingen ; Jochen Triesch
COMMENTS:       Submitted to IEEE Transactions on Neural Networks and Learning Systems
HIGHLIGHT:      To overcome these challenges, we propose a new method for training a deep neural network that distills particularly representative training examples and augments the training data by mixing these samples from one class with those from the same and other classes to create additional training samples.

9, TITLE:       Fair DARTS: Eliminating Unfair Advantages in Differentiable Architecture Search
http://arxiv.org/abs/1911.12126
AUTHORS:        Xiangxiang Chu ; Tianbao Zhou ; Bo Zhang ; Jixiang Li
COMMENTS:       Accepted to ECCV 2020, camera ready version
HIGHLIGHT:      Thereby, we present a novel approach called Fair DARTS where the exclusive competition is relaxed to be collaborative.

10, TITLE:      A Causal Linear Model to Quantify Edge Unfairness for Unfair Edge Prioritization and Discrimination Removal
http://arxiv.org/abs/2007.05516
AUTHORS:        Pavan Ravishankar ; Pranshu Malviya ; Balaraman Ravindran
COMMENTS:       Accepted in the Workshop on Law and Machine Learning, ICML 2020; First two authors contributed equally
HIGHLIGHT:      Prior work of (Zhang, et al., 2017) identifies and removes discrimination after data is generated but does not suggest a methodology to mitigate unfairness in the data generation phase.

11, TITLE:      Learn to Earn: Enabling Coordination within a Ride Hailing Fleet
http://arxiv.org/abs/2006.10904
AUTHORS:        Harshal A. Chaudhari ; John W. Byers ; Evimaria Terzi
COMMENTS:       16 pages, 9 figures
HIGHLIGHT:      An ideal solution aims to minimize the response time for each hyper local passenger ride request, while simultaneously maintaining high demand satisfaction and supply utilization across the entire city.

12, TITLE:      Cross-Task Transfer for Geotagged Audiovisual Aerial Scene Recognition
http://arxiv.org/abs/2005.08449
AUTHORS:        Di Hu ; Xuhong Li ; Lichao Mou ; Pu Jin ; Dong Chen ; Liping Jing ; Xiaoxiang Zhu ; Dejing Dou
COMMENTS:       ECCV 2020
HIGHLIGHT:      Inspired by the multi-channel perception theory in cognition science, in this paper, for improving the performance on the aerial scene recognition, we explore a novel audiovisual aerial scene recognition task using both images and sounds as input. For this purpose, we have constructed a new dataset named AuDio Visual Aerial sceNe reCognition datasEt (ADVANCE).

13, TITLE:      Anchor & Transform: Learning Sparse Representations of Discrete Objects
http://arxiv.org/abs/2003.08197
AUTHORS:        Paul Pu Liang ; Manzil Zaheer ; Yuan Wang ; Amr Ahmed
HIGHLIGHT:      In this paper, we design an efficient embedding algorithm that learns a small set of anchor embeddings and a sparse transformation matrix.

14, TITLE:      A review: Deep learning for medical image segmentation using multi-modality fusion
http://arxiv.org/abs/2004.10664
AUTHORS:        Tongxue Zhou ; Su Ruan ; Stéphane Canu
COMMENTS:       26 pages, 8 figures
HIGHLIGHT:      In this paper, we give an overview of deep learning-based approaches for multi-modal medical image segmentation task.

15, TITLE:      Few-shot Compositional Font Generation with Dual Memory
http://arxiv.org/abs/2005.10510
AUTHORS:        Junbum Cha ; Sanghyuk Chun ; Gayoung Lee ; Bado Lee ; Seonghyeon Kim ; Hwalsuk Lee
COMMENTS:       ECCV 2020 camera-ready
HIGHLIGHT:      In this paper, we focus on compositional scripts, a widely used letter system in the world, where each glyph can be decomposed by several components.

16, TITLE:      Design and Interpretation of Universal Adversarial Patches in Face Detection
http://arxiv.org/abs/1912.05021
AUTHORS:        Xiao Yang ; Fangyun Wei ; Hongyang Zhang ; Jun Zhu
HIGHLIGHT:      We propose new optimization-based approaches to automatic design of universal adversarial patches for varying goals of the attack, including scenarios in which true positives are suppressed without introducing false positives.

17, TITLE:      Modified Possibilistic Fuzzy C-Means Algorithm for Clustering Incomplete Data Sets
http://arxiv.org/abs/2007.04908
AUTHORS:         Rustam ; Koredianto Usman ; Mudyawati Kamaruddin ; Dina Chamidah ;  Nopendri ; Khaerudin Saleh ; Yulinda Eliskar ; Ismail Marzuki
COMMENTS:       13 pages, 13 figures, submitted to Acta Polytechnica as scientific journal published by the Czech Technical University in Prague
HIGHLIGHT:      Therefore, in this study, we propose a modification of the PFCM algorithm that can be applied to incomplete data sets clustering.

18, TITLE:      Rethinking Anticipation Tasks: Uncertainty-aware Anticipation of Sparse Surgical Instrument Usage for Context-aware Assistance
http://arxiv.org/abs/2007.00548
AUTHORS:        Dominik Rivoir ; Sebastian Bodenstedt ; Isabel Funke ; Felix von Bechtolsheim ; Marius Distler ; Jürgen Weitz ; Stefanie Speidel
COMMENTS:       Accepted at MICCAI 2020
HIGHLIGHT:      We propose a novel learning task for anticipation of instrument usage in laparoscopic videos that overcomes these limitations.

19, TITLE:      Hierarchical Dynamic Filtering Network for RGB-D Salient Object Detection
http://arxiv.org/abs/2007.06227
AUTHORS:        Youwei Pang ; Lihe Zhang ; Xiaoqi Zhao ; Huchuan Lu
COMMENTS:       Accepted by ECCV 2020
HIGHLIGHT:      In this paper, we explore these issues from a new perspective.

20, TITLE:      Unpaired Image-to-Image Translation using Adversarial Consistency Loss
http://arxiv.org/abs/2003.04858
AUTHORS:        Yihao Zhao ; Ruihai Wu ; Hao Dong
HIGHLIGHT:      In this paper, we propose a novel adversarial-consistency loss for image-to-image translation.

21, TITLE:      Deep Learning for Abstract Argumentation Semantics
http://arxiv.org/abs/2007.07629
AUTHORS:        Dennis Craandijk ; Floris Bex
COMMENTS:       Accepted at the main track of IJCAI 2020. SOLE copyright holder is IJCAI (international Joint Conferences on Artificial Intelligence)
HIGHLIGHT:      In this paper, we present a learning-based approach to determining acceptance of arguments under several abstract argumentation semantics.

22, TITLE:      Bottom-Up Temporal Action Localization with Mutual Regularization
http://arxiv.org/abs/2002.07358
AUTHORS:        Peisen Zhao ; Lingxi Xie ; Chen Ju ; Ya Zhang ; Yanfeng Wang ; Qi Tian
COMMENTS:       Accepted by ECCV2020
HIGHLIGHT:      To alleviate this problem, we introduce two regularization terms to mutually regularize the learning procedure: the Intra-phase Consistency (IntraC) regularization is proposed to make the predictions verified inside each phase; and the Inter-phase Consistency (InterC) regularization is proposed to keep consistency between these phases.

23, TITLE:      Augmenting Visual Place Recognition with Structural Cues
http://arxiv.org/abs/2003.00278
AUTHORS:        Amadeus Oertel ; Titus Cieslewski ; Davide Scaramuzza
COMMENTS:       8 pages, published in RA-L & IROS 2020
HIGHLIGHT:      In this paper, we propose to augment image-based place recognition with structural cues.

24, TITLE:      CURL: Neural Curve Layers for Global Image Enhancement
http://arxiv.org/abs/1911.13175
AUTHORS:        Sean Moran ; Steven McDonagh ; Gregory Slabaugh
HIGHLIGHT:      We present a novel approach to adjust global image properties such as colour, saturation, and luminance using human-interpretable image enhancement curves, inspired by the Photoshop curves tool.

25, TITLE:      CNN based Road User Detection using the 3D Radar Cube
http://arxiv.org/abs/2004.12165
AUTHORS:        Andras Palffy ; Jiaao Dong ; Julian F. P. Kooij ; Dariu M. Gavrila
HIGHLIGHT:      This letter presents a novel radar based, single-frame, multi-class detection method for moving road users (pedestrian, cyclist, car), which utilizes low-level radar cube data.

26, TITLE:      Improved object recognition using neural networks trained to mimic the brain's statistical properties
http://arxiv.org/abs/1905.10679
AUTHORS:        Callie Federer ; Haoyan Xu ; Alona Fyshe ; Joel Zylberberg
HIGHLIGHT:      Improved object recognition using neural networks trained to mimic the brain's statistical properties

27, TITLE:      Learning to Transfer Learn: Reinforcement Learning-Based Selection for Adaptive Transfer Learning
http://arxiv.org/abs/1908.11406
AUTHORS:        Linchao Zhu ; Sercan O. Arik ; Yi Yang ; Tomas Pfister
HIGHLIGHT:      We propose a novel adaptive transfer learning framework, learning to transfer learn (L2TL), to improve performance on a target dataset by careful extraction of the related information from a source dataset.

28, TITLE:      Patch-wise Attack for Fooling Deep Neural Network
http://arxiv.org/abs/2007.06765
AUTHORS:        Lianli Gao ; Qilong Zhang ; Jingkuan Song ; Xianglong Liu ; Heng Tao Shen
COMMENTS:       Accepted by ECCV 2020
HIGHLIGHT:      Motivated by this, we propose a patch-wise iterative algorithm -- a black-box attack towards mainstream normally trained and defense models, which differs from the existing attack methods manipulating pixel-wise noise.

29, TITLE:      ATSO: Asynchronous Teacher-Student Optimization for Semi-Supervised Medical Image Segmentation
http://arxiv.org/abs/2006.13461
AUTHORS:        Xinyue Huo ; Lingxi Xie ; Jianzhong He ; Zijie Yang ; Qi Tian
HIGHLIGHT:      To alleviate this issue, we propose ATSO, an asynchronous version of teacher-student optimization.

30, TITLE:      Train Your Data Processor: Distribution-Aware and Error-Compensation Coordinate Decoding for Human Pose Estimation
http://arxiv.org/abs/2007.05887
AUTHORS:        Feiyu Yang ; Zhan Song ; Zhenzhong Xiao ; Yu Chen ; Zhe Pan ; Min Zhang ; Min Xue ; Yaoyang Mo ; Yao Zhang ; Guoxiong Guan ; Beibei Qian
COMMENTS:       Improve the state-of-the-art of COCO keypoint detection challenge by 1-2 AP. Project page: https://github.com/fyang235/DAEC
HIGHLIGHT:      Serving as a model-agnostic plug-in, DAEC learns its decoding strategy from training data and remarkably improves the performance of a variety of state-of-the-art human pose estimation models with negligible extra computation.

31, TITLE:      The Best of Both Modes: Separately Leveraging RGB and Depth for Unseen Object Instance Segmentation
http://arxiv.org/abs/1907.13236
AUTHORS:        Christopher Xie ; Yu Xiang ; Arsalan Mousavian ; Dieter Fox
HIGHLIGHT:      We propose a novel method that separately leverages synthetic RGB and synthetic depth for unseen object instance segmentation. To train our method, we introduce a large-scale synthetic dataset of random objects on tabletops.

32, TITLE:      Object Segmentation Tracking from Generic Video Cues
http://arxiv.org/abs/1910.02258
AUTHORS:        Amirhossein Kardoost ; Sabine Müller ; Joachim Weickert ; Margret Keuper
HIGHLIGHT:      We propose a light-weight variational framework for online tracking of object segmentations in videos based on optical flow and image boundaries.

33, TITLE:      A Finite Time Combinatorial Algorithm for Instantaneous Dynamic Equilibrium Flows
http://arxiv.org/abs/2007.07808
AUTHORS:        Lukas Graf ; Tobias Harks
COMMENTS:       26 pages, 11 figures
HIGHLIGHT:      We analyze IDE within the Vickrey bottleneck model, where current travel times along a path consist of the physical travel times plus the sum of waiting times in all the queues along a path.

34, TITLE:      6D Camera Relocalization in Ambiguous Scenes via Continuous Multimodal Inference
http://arxiv.org/abs/2004.04807
AUTHORS:        Mai Bui ; Tolga Birdal ; Haowen Deng ; Shadi Albarqouni ; Leonidas Guibas ; Slobodan Ilic ; Nassir Navab
COMMENTS:       Accepted for publication at ECCV 2020. Project page under https://multimodal3dvision.github.io
HIGHLIGHT:      We present a multimodal camera relocalization framework that captures ambiguities and uncertainties with continuous mixture models defined on the manifold of camera poses. We introduce a new dataset specifically designed to foster camera localization research in ambiguous environments and exhaustively evaluate our method on synthetic as well as real data on both ambiguous scenes and on non-ambiguous benchmark datasets.

35, TITLE:      Adversarial Robustness Assessment: Why both $L_0$ and $L_\infty$ Attacks Are Necessary
http://arxiv.org/abs/1906.06026
AUTHORS:        Shashank Kotyan ; Danilo Vasconcellos Vargas
HIGHLIGHT:      Based on this, we propose a model agnostic dual quality assessment method, together with the concept of robustness levels to tackle them.

36, TITLE:      Image Processing and Quality Control for Abdominal Magnetic Resonance Imaging in the UK Biobank
http://arxiv.org/abs/2007.01251
AUTHORS:        Nicolas Basty ; Yi Liu ; Madeleine Cule ; E. Louise Thomas ; Jimmy D. Bell ; Brandon Whitcher
COMMENTS:       Fixed 2 references
HIGHLIGHT:      Detection of fat-water swaps in the Dixon series is performed by a deep learning model and corrected automatically.

37, TITLE:      On Scaling Data-Driven Loop Invariant Inference
http://arxiv.org/abs/1911.11728
AUTHORS:        Sahil Bhatia ; Saswat Padhi ; Nagarajan Natarajan ; Rahul Sharma ; Prateek Jain
HIGHLIGHT:      In this paper, we study these scalability issues and address them in our tool oasis that improves the scale of data-driven invariant inference and outperforms state-of-the-art systems on benchmarks from the invariant inference track of the Syntax Guided Synthesis competition.

38, TITLE:      Efficiently Calibrating Cable-Driven Surgical Robots with RGBD Fiducial Sensing and Recurrent Neural Networks
http://arxiv.org/abs/2003.08520
AUTHORS:        Minho Hwang ; Brijen Thananjeyan ; Samuel Paradis ; Daniel Seita ; Jeffrey Ichnowski ; Danyal Fer ; Thomas Low ; Ken Goldberg
COMMENTS:       8 pages, 11 figures, 3 tables
HIGHLIGHT:      We propose a novel approach to efficiently calibrate such robots by placing a 3D printed fiducial coordinate frames on the arm and end-effector that is tracked using RGBD sensing.

39, TITLE:      Divide, Conquer, and Combine: a New Inference Strategy for Probabilistic Programs with Stochastic Support
http://arxiv.org/abs/1910.13324
AUTHORS:        Yuan Zhou ; Hongseok Yang ; Yee Whye Teh ; Tom Rainforth
COMMENTS:       Published at the 37th International Conference on Machine Learning (ICML 2020)
HIGHLIGHT:      To address this, we introduce a new inference framework: Divide, Conquer, and Combine, which remains efficient for such models, and show how it can be implemented as an automated and generic PPS inference engine.

40, TITLE:      REPrune: Filter Pruning via Representative Election
http://arxiv.org/abs/2007.06932
AUTHORS:        Mincheol Park ; Woojeong Kim ; Suhyun Kim
COMMENTS:       Under Review at ECCV 2020
HIGHLIGHT:      Our novel pruning method entitled "REPrune" addresses this problem by selecting representative filters via clustering.

41, TITLE:      Smooth Points on Semi-algebraic Sets
http://arxiv.org/abs/2002.04707
AUTHORS:        Katherine Harris ; Jonathan D. Hauenstein ; Agnes Szanto
HIGHLIGHT:      In this paper, we present a simple algorithm based on computing the critical points of some well-chosen function that guarantees the computation of smooth points in each connected compact component of a real (semi)-algebraic set.

42, TITLE:      Globally Optimal Segmentation of Mutually Interacting Surfaces using Deep Learning
http://arxiv.org/abs/2007.01259
AUTHORS:        Hui Xie ; Zhe Pan ; Leixin Zhou ; Fahim A Zaman ; Danny Chen ; Jost B Jonas ; Yaxing Wang ; Xiaodong Wu
COMMENTS:       11 pages main content and reference, plus 10 pages appendix, total 21 pages
HIGHLIGHT:      In this work, we propose to parameterize the surface cost functions in the graph model and leverage DL to learn those parameters.

43, TITLE:      Structured Landmark Detection via Topology-Adapting Deep Graph Learning
http://arxiv.org/abs/2004.08190
AUTHORS:        Weijian Li ; Yuhang Lu ; Kang Zheng ; Haofu Liao ; Chihung Lin ; Jiebo Luo ; Chi-Tung Cheng ; Jing Xiao ; Le Lu ; Chang-Fu Kuo ; Shun Miao
COMMENTS:       Accepted to ECCV-20. Camera-ready with supplementary material
HIGHLIGHT:      In this work, we present a new topology-adapting deep graph learning approach for accurate anatomical facial and medical (e.g., hand, pelvis) landmark detection.

44, TITLE:      AutoSTR: Efficient Backbone Search for Scene Text Recognition
http://arxiv.org/abs/2003.06567
AUTHORS:        Hui Zhang ; Quanming Yao ; Mingkun Yang ; Yongchao Xu ; Xiang Bai
COMMENTS:       ECCV 2020
HIGHLIGHT:      In this work, inspired by the success of neural architecture search (NAS), which can identify better architectures than human-designed ones, we propose automated STR (AutoSTR) to search data-dependent backbones to boost text recognition performance.

45, TITLE:      Geometric Attention for Prediction of Differential Properties in 3D Point Clouds
http://arxiv.org/abs/2007.02571
AUTHORS:        Albert Matveev ; Alexey Artemov ; Denis Zorin ; Evgeny Burnaev
HIGHLIGHT:      In this study, we present a geometric attention mechanism that can provide such properties in a learnable fashion.

46, TITLE:      Complex Markov Logic Networks: Expressivity and Liftability
http://arxiv.org/abs/2002.10259
AUTHORS:        Ondrej Kuzelka
COMMENTS:       Fixed typos in Lemma 1 and Section 7. Paper accepted to UAI 2020
HIGHLIGHT:      We introduce complex MLNs, which use complex-valued weights, and we show that, unlike standard MLNs with real-valued weights, complex MLNs are fully expressive.

47, TITLE:      PointAR: Efficient Lighting Estimation for Mobile Augmented Reality
http://arxiv.org/abs/2004.00006
AUTHORS:        Yiqin Zhao ; Tian Guo
HIGHLIGHT:      We propose an efficient lighting estimation pipeline that is suitable to run on modern mobile devices, with comparable resource complexities to state-of-the-art mobile deep learning models.

48, TITLE:      FakeSpotter: A Simple yet Robust Baseline for Spotting AI-Synthesized Fake Faces
http://arxiv.org/abs/1909.06122
AUTHORS:        Run Wang ; Felix Juefei-Xu ; Lei Ma ; Xiaofei Xie ; Yihao Huang ; Jian Wang ; Yang Liu
COMMENTS:       Accepted to IJCAI 2020; SOLE copyright holder is IJCAI (international Joint Conferences on Artificial Intelligence), all rights reserved. https://www.ijcai.org/Proceedings/2020/333
HIGHLIGHT:      In this work, we propose a novel approach, named FakeSpotter, based on monitoring neuron behaviors to spot AI-synthesized fake faces.

49, TITLE:      Span-ConveRT: Few-shot Span Extraction for Dialog with Pretrained Conversational Representations
http://arxiv.org/abs/2005.08866
AUTHORS:        Sam Coope ; Tyler Farghly ; Daniela Gerz ; Ivan Vulić ; Matthew Henderson
COMMENTS:       ACL 2020 (updated version with errata)
HIGHLIGHT:      We introduce Span-ConveRT, a light-weight model for dialog slot-filling which frames the task as a turn-based span extraction task.

50, TITLE:      Localising Faster: Efficient and precise lidar-based robot localisation in large-scale environments
http://arxiv.org/abs/2003.01875
AUTHORS:        Li Sun ; Daniel Adolfsson ; Martin Magnusson ; Henrik Andreasson ; Ingmar Posner ; Tom Duckett
COMMENTS:       7 pages, 5 pages. Accepted by IEEE International Conference on Robotics and Automation (ICRA) 2020
HIGHLIGHT:      This paper proposes a novel approach for global localisation of mobile robots in large-scale environments.

51, TITLE:      Explaining Deep Neural Networks using Unsupervised Clustering
http://arxiv.org/abs/2007.07477
AUTHORS:        Yu-han Liu ; Sercan O. Arik
HIGHLIGHT:      We propose a novel method to explain trained deep neural networks (DNNs), by distilling them into surrogate models using unsupervised clustering.

52, TITLE:      Towards Robust Learning with Different Label Noise Distributions
http://arxiv.org/abs/1912.08741
AUTHORS:        Diego Ortego ; Eric Arazo ; Paul Albert ; Noel E. O'Connor ; Kevin McGuinness
HIGHLIGHT:      SSL is then applied twice, once to improve the clean-noisy detection and again for training the final model.

53, TITLE:      In-Domain GAN Inversion for Real Image Editing
http://arxiv.org/abs/2004.00049
AUTHORS:        Jiapeng Zhu ; Yujun Shen ; Deli Zhao ; Bolei Zhou
COMMENTS:       ECCV 2020 camera-ready
HIGHLIGHT:      To solve this problem, we propose an in-domain GAN inversion approach, which not only faithfully reconstructs the input image but also ensures the inverted code to be semantically meaningful for editing.

54, TITLE:      Flowtron: an Autoregressive Flow-based Generative Network for Text-to-Speech Synthesis
http://arxiv.org/abs/2005.05957
AUTHORS:        Rafael Valle ; Kevin Shih ; Ryan Prenger ; Bryan Catanzaro
COMMENTS:       10 pages, 7 pictures
HIGHLIGHT:      In this paper we propose Flowtron: an autoregressive flow-based generative network for text-to-speech synthesis with control over speech variation and style transfer.

55, TITLE:      FoveaBox: Beyond Anchor-based Object Detector
http://arxiv.org/abs/1904.03797
AUTHORS:        Tao Kong ; Fuchun Sun ; Huaping Liu ; Yuning Jiang ; Lei Li ; Jianbo Shi
COMMENTS:       IEEE Transactions on Image Processing, code at: https://github.com/taokong/FoveaBox
HIGHLIGHT:      We present FoveaBox, an accurate, flexible, and completely anchor-free framework for object detection.

56, TITLE:      Efficient Deep Neural Network for Photo-realistic Image Super-Resolution
http://arxiv.org/abs/1903.02240
AUTHORS:        Namhyuk Ahn ; Byungkon Kang ; Kyung-Ah Sohn
HIGHLIGHT:      To facilitate the use of a deep model under such demands, we focus on keeping the network efficient while maintaining its performance.

57, TITLE:      Label Efficient Visual Abstractions for Autonomous Driving
http://arxiv.org/abs/2005.10091
AUTHORS:        Aseem Behl ; Kashyap Chitta ; Aditya Prakash ; Eshed Ohn-Bar ; Andreas Geiger
COMMENTS:       International Conference on Intelligent Robots and Systems (IROS), 2020. First two authors contributed equally, listed in alphabetical order
HIGHLIGHT:      In this work, we seek to quantify the impact of reducing segmentation annotation costs on learned behavior cloning agents.

58, TITLE:      Context Matters: Recovering Human Semantic Structure from Machine Learning Analysis of Large-Scale Text Corpora
http://arxiv.org/abs/1910.06954
AUTHORS:        Marius Cătălin Iordan ; Tyler Giallanza ; Cameron T. Ellis ; Nicole M. Beckage ; Jonathan D. Cohen
COMMENTS:       Main Text: 35 pages, 5 figures; Supplemental: 21 pages, 11 figures, 6 tables
HIGHLIGHT:      Here, we introduce a novel approach of generating embeddings motivated by the psychological theory that semantic context plays a critical role in human judgments.

59, TITLE:      An Uncertainty-based Human-in-the-loop System for Industrial Tool Wear Analysis
http://arxiv.org/abs/2007.07129
AUTHORS:        Alexander Treiss ; Jannis Walk ; Niklas Kühl
COMMENTS:       Alexander Treiss and Jannis Walk contributed equally in shared first authorship. To be published at ECML-PKDD 2020
HIGHLIGHT:      To address these issues, we use uncertainty measures based on Monte-Carlo dropout in the context of a human-in-the-loop system to increase the system's transparency and performance.

60, TITLE:      Actor-Context-Actor Relation Network for Spatio-Temporal Action Localization
http://arxiv.org/abs/2006.07976
AUTHORS:        Junting Pan ; Siyu Chen ; Zheng Shou ; Jing Shao ; Hongsheng Li
COMMENTS:       1st place solution in ActivityNet Challenge 2020 -- AVA-Kinetics Task
HIGHLIGHT:      In this paper, we propose to explicitly model the Actor-Context-Actor Relation, which can capture indirect high-order supportive information for effectively reasoning actors' actions in complex scenes.

61, TITLE:      Generative Latent Implicit Conditional Optimization when Learning from Small Sample
http://arxiv.org/abs/2003.14297
AUTHORS:        Idan Azuri ; Daphna Weinshall
COMMENTS:       8 pages, 7 figures
HIGHLIGHT:      In this paper we propose a novel such method called GLICO (Generative Latent Implicit Conditional Optimization).

62, TITLE:      Learning Object Permanence from Video
http://arxiv.org/abs/2003.10469
AUTHORS:        Aviv Shamsian ; Ofri Kleinfeld ; Amir Globerson ; Gal Chechik
COMMENTS:       16th European Conference on Computer Vision (ECCV 2020)
HIGHLIGHT:      Here we introduce the setup of learning Object Permanence from data.

63, TITLE:      Representation Quality Of Neural Networks Links To Adversarial Attacks and Defences
http://arxiv.org/abs/1906.06627
AUTHORS:        Shashank Kotyan ; Danilo Vasconcellos Vargas ; Moe Matsuki
HIGHLIGHT:      Here, we propose a method to understand the representation quality of the neural networks using a novel test based on Zero-Shot Learning, entitled Raw Zero-Shot.

64, TITLE:      A Balanced and Uncertainty-aware Approach for Partial Domain Adaptation
http://arxiv.org/abs/2003.02541
AUTHORS:        Jian Liang ; Yunbo Wang ; Dapeng Hu ; Ran He ; Jiashi Feng
COMMENTS:       Accepted to ECCV 2020
HIGHLIGHT:      In this paper, we build on domain adversarial learning and propose a novel domain adaptation method BA$^3$US with two new techniques termed Balanced Adversarial Alignment (BAA) and Adaptive Uncertainty Suppression (AUS), respectively.

65, TITLE:      CenterNet3D:An Anchor free Object Detector for Autonomous Driving
http://arxiv.org/abs/2007.07214
AUTHORS:        Guojun Wang ; Bin Tian ; Yunfeng Ai ; Tong Xu ; Long Chen ; Dongpu Cao
COMMENTS:       9 pages, 3 figures
HIGHLIGHT:      In this paper, we eliminate anchors and model an object as a single point the center point of its bounding box.

66, TITLE:      Implicit Diversity in Image Summarization
http://arxiv.org/abs/1901.10265
AUTHORS:        L. Elisa Celis ; Vijay Keswani
HIGHLIGHT:      We develop a novel approach that takes as input a visibly diverse control set of images and uses this set to select a set of images of people in response to a query.

67, TITLE:      Graph-Based Social Relation Reasoning
http://arxiv.org/abs/2007.07453
AUTHORS:        Wanhua Li ; Yueqi Duan ; Jiwen Lu ; Jianjiang Feng ; Jie Zhou
COMMENTS:       ECCV 2020
HIGHLIGHT:      In this paper, we propose a simpler, faster, and more accurate method named graph relational reasoning network (GR2N) for social relation recognition.

68, TITLE:      Facial Expression Recognition using Facial Landmark Detection and Feature Extraction via Neural Networks
http://arxiv.org/abs/1812.04510
AUTHORS:        Fuzail Khan
HIGHLIGHT:      The proposed framework in this paper has the primary objective of classifying the facial expression shown by a person.

69, TITLE:      Lifted Inference in 2-Variable Markov Logic Networks with Function and Cardinality Constraints Using Discrete Fourier Transform
http://arxiv.org/abs/2006.03432
AUTHORS:        Ondrej Kuzelka
COMMENTS:       arXiv admin note: text overlap with arXiv:2002.10259, This version: fixed a typo in Section 3.1
HIGHLIGHT:      In this paper we show that inference in 2-variable Markov logic networks (MLNs) with cardinality and function constraints is domain-liftable.

70, TITLE:      Looking back to lower-level information in few-shot learning
http://arxiv.org/abs/2005.13638
AUTHORS:        Zhongjie Yu ; Sebastian Raschka
COMMENTS:       13 pages, 2 figures; fixed typographic errors and added journal ref
HIGHLIGHT:      In this work, we propose the utilization of lower-level, supporting information, namely the feature embeddings of the hidden neural network layers, to improve classifier accuracy.

71, TITLE:      TraceCaps: A Capsule-based Neural Network for Semantic Segmentation
http://arxiv.org/abs/1901.02920
AUTHORS:        Tao Sun ; Zhewei Wang ; C. D. Smith ; Jundong Liu
HIGHLIGHT:      In this paper, we propose a capsule-based neural network model to solve the semantic segmentation problem.

72, TITLE:      VQA-LOL: Visual Question Answering under the Lens of Logic
http://arxiv.org/abs/2002.08325
AUTHORS:        Tejas Gokhale ; Pratyay Banerjee ; Chitta Baral ; Yezhou Yang
COMMENTS:       Accepted to ECCV 2020
HIGHLIGHT:      In this paper, we investigate whether visual question answering (VQA) systems trained to answer a question about an image, are able to answer the logical composition of multiple such questions.

73, TITLE:      Cascade Network with Guided Loss and Hybrid Attention for Two-view Geometry
http://arxiv.org/abs/2007.05706
AUTHORS:        Zhi Chen ; Fan Yang ; Wenbing Tao
HIGHLIGHT:      In this paper, we are committed to designing a high-performance network for two-view geometry.

74, TITLE:      PT2PC: Learning to Generate 3D Point Cloud Shapes from Part Tree Conditions
http://arxiv.org/abs/2003.08624
AUTHORS:        Kaichun Mo ; He Wang ; Xinchen Yan ; Leonidas J. Guibas
COMMENTS:       ECCV 2020
HIGHLIGHT:      In order to learn such a conditional shape generation procedure in an end-to-end fashion, we propose a conditional GAN "part tree"-to-"point cloud" model (PT2PC) that disentangles the structural and geometric factors.

75, TITLE:      Benchmarking Differentially Private Residual Networks for Medical Imagery
http://arxiv.org/abs/2005.13099
AUTHORS:        Sahib Singh ; Harshvardhan Sikka ; Sasikanth Kotti ; Andrew Trask
COMMENTS:       5 Pages, 4 Figures
HIGHLIGHT:      In this paper we measure the effectiveness of $\epsilon$-Differential Privacy (DP) when applied to medical imaging.

76, TITLE:      On Box-Cox Transformation for Image Normality and Pattern Classification
http://arxiv.org/abs/2004.07210
AUTHORS:        Abbas Cheddad
COMMENTS:       The paper has 4 Tables and 6 Figures
HIGHLIGHT:      We compare the effect of this light-weight Box-Cox transformation with well-established state-of-the-art low light image enhancement techniques.

77, TITLE:      Quantization Guided JPEG Artifact Correction
http://arxiv.org/abs/2004.09320
AUTHORS:        Max Ehrlich ; Larry Davis ; Ser-Nam Lim ; Abhinav Shrivastava
COMMENTS:       Published in the proceedings of ECCV 2020, please see our released code and models at https://gitlab.com/Queuecumber/quantization-guided-ac
HIGHLIGHT:      We solve this problem by creating a novel architecture which is parameterized by the JPEG files quantization matrix.

78, TITLE:      Angle-based Search Space Shrinking for Neural Architecture Search
http://arxiv.org/abs/2004.13431
AUTHORS:        Yiming Hu ; Yuding Liang ; Zichao Guo ; Ruosi Wan ; Xiangyu Zhang ; Yichen Wei ; Qingyi Gu ; Jian Sun
COMMENTS:       Accepted in ECCV 2020
HIGHLIGHT:      In this work, we present a simple and general search space shrinking method, called Angle-Based search space Shrinking (ABS), for Neural Architecture Search (NAS).

79, TITLE:      Learning Where to Focus for Efficient Video Object Detection
http://arxiv.org/abs/1911.05253
AUTHORS:        Zhengkai Jiang ; Yu Liu ; Ceyuan Yang ; Jihao Liu ; Peng Gao ; Qian Zhang ; Shiming Xiang ; Chunhong Pan
COMMENTS:       Accepted to ECCV 2020
HIGHLIGHT:      Previous approaches exploit to propagate and aggregate features across video frames by using optical flow-warping.

80, TITLE:      AdvPC: Transferable Adversarial Perturbations on 3D Point Clouds
http://arxiv.org/abs/1912.00461
AUTHORS:        Abdullah Hamdi ; Sara Rojas ; Ali Thabet ; Bernard Ghanem
COMMENTS:       Presented at European conference on computer vision (ECCV), 2020. The code is available at https://github.com/ajhamdi/AdvPC
HIGHLIGHT:      In this work, we present novel data-driven adversarial attacks against 3D point cloud networks.

81, TITLE:      Learning Depth With Very Sparse Supervision
http://arxiv.org/abs/2003.00752
AUTHORS:        Antonio Loquercio ; Alexey Dosovitskiy ; Davide Scaramuzza
COMMENTS:       Accepted for Publication at the IEEE Robotics and Automation Letters (RA-L) 2020, and International Conference on Intelligent Robots and Systems (IROS) 2020
HIGHLIGHT:      This paper explores a new approach to learning depth perception requiring neither of those.

82, TITLE:      On Emergent Communication in Competitive Multi-Agent Teams
http://arxiv.org/abs/2003.01848
AUTHORS:        Paul Pu Liang ; Jeffrey Chen ; Ruslan Salakhutdinov ; Louis-Philippe Morency ; Satwik Kottur
COMMENTS:       AAMAS 2020, code: https://github.com/pliang279/Competitive-Emergent-Communication
HIGHLIGHT:      In this work, we investigate whether competition for performance from an external, similar agent team could act as a social influence that encourages multi-agent populations to develop better communication protocols for improved performance, compositionality, and convergence speed.