Vision Darknet Detect

Autoware package based on Darknet that supports Yolov3 and Yolov2 image detector.

Requirements

• NVIDIA GPU with CUDA installed
• Pretrained YOLOv3 or YOLOv2 model on COCO dataset, Models found on the YOLO website.
• The weights file must be placed in vision_darknet_detect/darknet/data/.

How to launch

• From a sourced terminal:

  • roslaunch vision_darknet_detect vision_yolo3_detect.launch
  • roslaunch vision_darknet_detect vision_yolo2_detect.launch

• From Runtime Manager:

Computing Tab -> Detection/ vision_detector -> vision_darknet_detect You can change the config and weights file, as well as other parameters, by clicking [app]

Parameters

Launch file available parameters:

Subscribed topics

Published topics

Video

Changelog for package vision_yolo3_detect

1.11.0 (2019-03-21)

• Removing CUDA dependencies for Darknet Yolov3 (#1784)

  * Removing CUDA dependencies for Darknet yolov3 If the host machine does not have CUDA, this will build the vision_darknet_detect package based on a pre-built darknet directory (which doesn't require CUDA as there are no CUDA dependencies for yolov3).

  * Update ros/src/computing/perception/detection/vision_detector/packages/vision_darknet_detect/CMakeLists.txt Co-Authored-By: K1504296 <<greytrt@gmail.com>>

• Fix license notice in corresponding package.xml

• Initial release of object filter

• Contributors: Abraham Monrroy, Theodore, amc-nu

1.10.0 (2019-01-17)

• Fixes for catkin_make
• [fix] SSD detector, cmake colcon (#1837)

  • Fixes for new colcon script on ssd cuda based node

  • Fixed to RTM and darknet launch files

  • catkin_fix

  • • catkin & colcon build successfully
  • reverted back run to devel space (for the time being)
• Switch to Apache 2 license (develop branch) (#1741)
  • Switch to Apache 2

  * Replace BSD-3 license header with Apache 2 and reassign copyright to the Autoware Foundation.

  • Update license on Python files
  • Update copyright years
  • Add #ifndef/define _POINTS_IMAGE_H_
  • Updated license comment
• Use colcon as the build tool (#1704)
  • Switch to colcon as the build tool instead of catkin
  • Added cmake-target
  • Added note about the second colcon call
  • Added warning about catkin* scripts being deprecated
  • Fix COLCON_OPTS
  • Added install targets
  • Update Docker image tags
  • Message packages fixes
  • Fix missing dependency
• Feature/perception visualization cleanup (#1648)
  • • Initial commit for visualization package

  • Removal of all visualization messages from perception nodes

  • Visualization dependency removal

  • Launch file modification

  • • Fixes to visualization

  • Error on Clustering CPU

  • Reduce verbosity on markers

  • intial commit

  • • Changed to 2 spaces indentation

  • Added README

  • Fixed README messages type

  • 2 space indenting

  • ros clang format

  • Publish acceleration and velocity from ukf tracker

  • Remove hardcoded path

  • Updated README

  • updated prototype

  • Prototype update for header and usage

  • Removed unknown label from being reported

  • Updated publishing orientation to match develop

  • • Published all the trackers

File truncated at 100 lines see the full file