Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add TensorRT YOLO implementation #17

Draft
wants to merge 16 commits into
base: master
Choose a base branch
from
Draft

Conversation

nickswalker
Copy link
Member

This branch needs to be cleaned up to pass CI.

Closes #5

JHLee0513 and others added 12 commits March 2, 2020 18:29
Wrap yolov3 submodule in a package, have the package use catkin python
setup to provide the module to scripts and other catkin packages
Paste camera demo with imports corrected
1. subscribes to image from sim
2. predicts object bounding boxes
3. publishes prediction
Includes a copy of Xinyu Wang's TensorRT implementation of the Ultralytics YOLOv3 implementation
Some headers headers shuffled out into the include directory so things build in a ROS environment
A bit of CMake setup to make sure things only attempt to build when there's CUDA on the machine
Change demo node to use plain model
Add readme with usage notes
Not touching the copy pasted code to make future diffing easier
Add clang files
@nickswalker nickswalker self-assigned this Aug 14, 2020
Fix whitespace, disable roslint
Use TensorRTs plugin options struct to pass in the number. Also copied
over some boilerplate for making other parameters configurable this way but
didn't hook them up fully
Tweak plugin registration point to match an update in upstream
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

Detect a known object
2 participants