Skip to content
/ TALL Public

TALL: Temporal Activity Localization via Language Query

Notifications You must be signed in to change notification settings

jiyanggao/TALL

Repository files navigation

TALL: Temporal Activity Localization via Language Query

This is the repository for our ICCV 2017 paper TALL: Temporal Activity Localization via Language Query.

Visual Features on TACoS

Download the C3D features for training set and test set of TACoS dataset. Modify the path to feature folders in main.py

Sentence Embeddings on TACoS

Download the Skip-thought sentence embeddings and sample files from here of TACoS Dataset, and put them under exp_data folder.

Reproduce the results on TACoS

python main.py

Charades-STA anno download

The sentence temporal annotations on Charades dataset are available here: train, test. The format is "[video name] [start time] [end time]##[sentence]". You may want to generate the skip-thought embeddings and C3D features on Charades-STA, and modify the codes slightly to reproduce the experiments.

Updates on Charades-STA performance

I did some anno cleaning for Charades-STA (compared to the version I used in ICCV paper), the updated performance is listed below. Please compare to these results when using Charades-STA.

Model R@1,IoU=0.5 R@1,IoU=0.7 R@5,IoU=0.5 R@5,IoU=0.7
CTRL (aln) 17.69 5.91 55.54 23.79
CTRL (reg-p) 19.22 6.64 57.98 25.22
CTRL (reg-np) 21.42 7.15 59.11 26.91

About

TALL: Temporal Activity Localization via Language Query

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages