Releases: opendatacam/opendatacam
Releases · opendatacam/opendatacam
v3.0.2
Highlighted Changes
- #427 Fixes issue with MongoDB where the latest MongoDB release (v5) does no longer run on Jetson devices.
- #420 Fixes issue in k8s setup
- #415 Fixes Docker compose installation instructions
- #356 Automatic creation of Docker Images
- #444 Fix Opendatacam Darknet Base docker tag
See Milestone v3.0.2 for a full list of issues and pull requests that went into the release.
Contributors
OpenDataCam is being developed by a small group of volunteers. 🙌
In this release we'd especially like to thank for their contribution (in alphabetic order) 🙏
v3.0.2-beta.2
Highlighted Changes
- #427 Fixes issue with MongoDB where the latest MongoDB release (v5) does no longer run on Jetson devices.
- #420 Fixes issue in k8s setup
- #415 Fixes Docker compose installation instructions
- #356 Automatic creation of Docker Images
- #444 Fix Opendatacam Darknet Base docker tag
See Milestone v3.0.2 for a full list of issues and pull requests that went into the release.
Contributors
OpenDataCam is being developed by a small group of volunteers. 🙌
In this release we'd especially like to thank for their contribution (in alphabetic order) 🙏
v3.0.2-beta.1
Highlighted Changes
- #427 Fixes issue with MongoDB where the latest MongoDB release (v5) does no longer run on Jetson devices.
- #420 Fixes issue in k8s setup
- #415 Fixes Docker compose installation instructions
- #356 Automatic creation of Docker Images
See Milestone v3.0.2 for a full list of issues and pull requests that went into the release.
Contributors
OpenDataCam is being developed by a small group of volunteers. 🙌
In this release we'd especially like to thank for their contribution (in alphabetic order) 🙏
v3.0.1
v3.0.0
New feature from v2.x
- Support for YOLOv4 and Yolov3-tiny-prn
- Drag & drop support for file, just drop a file on the UI and it will restart to process it
- Directional counters, see: #122 (comment)
- Fixed Memory leak on jetsons, expected running time should be counted in days now instead of hours ( #182 )
- Save / Reload counters config: see #122 (comment)
- Complete rewrite of Docker setup / install process in order to use docker-compose
Full changelog
- Improved performance by 6 FPS for Linux CUDA desktop machine by using cuDNN for the docker build
- Several smaller bugs fixes, clean up logs etc etc...
- Made ports configurable from config.json
- Add a configurable confidence threshold in config.json to filtrer out detections that are below this threshold. (default keep > 20% confidence)
- Add favicon
- Add "confirmation" modal when deleting recordings
- Add version number in side menu
- Updated all javascript dependencies: Next.js , Redux, React etc etc...
- Revert back to using the main ./darknet API as memory leak has been also fixed
- Update darknet fork once again using the new -dontdraw_bbox flag AlexeyAB kindly added for us : AlexeyAB/darknet@da05036
- Expose tracker config IOU and UNMATCHED FRAME tolerance
- Check if config.json version matches Opendatacam version
- Clean up code for drag& drop feature.. maybe add some UI tweaks like a drop indicator and a specific progress message
- Clean up and improve counter logic code
- Update API docs with new counters attributes
- Add Threshold to filter out large detections
- Update OpenMojis
- @fabianmoronzirfas added support for env variables: #222
- Add counter settings "minAngleWithCountingLine" : https://github.com/opendatacam/opendatacam/blob/development/documentation/CONFIG.md#counter-settings. Thanks @rantgithub for funding this feature
- Update documentation CONFIG.md as many default changed now.. we run on the demo.mp4 file now.. and file support is via drag & drop to simplify install process
- Update documentation Jetson nano
- Document how to run from custom weights with docker (possible now)
- Update Documentation for yolov3-tiny-prn and yolov4 weights
v3.0.0-beta.3
- Complete rewrite of Docker setup / install process in order to use
docker-compose
- Update darknet fork once again using the new -dontdraw_bbox flag AlexeyAB kindly added for us : AlexeyAB/darknet@da05036
- Expose tracker config IOU and UNMATCHED FRAME tolerance
- Check if config.json version matches Opendatacam version
- Clean up code for drag& drop feature.. maybe add some UI tweaks like a drop indicator and a specific progress message
- Clean up and improve counter logic code
- Update API docs with new counters attributes
- Add Threshold to filter out large detections
- Update OpenMojis
v3.0.0-beta.2
- Support for YOLOv4 (set
"NEURAL_NETWORK": "yolov4"
in config.json) - Upgraded docker setup and build workflow, see #178 , now you can run the docker container from jetson without needing to hack the support for CUDA.. It also simplified the Dockerfile file and made the build of OpenCV happening
- Result of previous point is support for RaspberryPi cam v2 (or any CSI cam compatible with nvarguscamerasrc) when running from the docker container for Jetson nano. Just set "VIDEO_INPUT": "raspberrycam"
- Improved performance by 6 FPS for Linux CUDA desktop machine by using cuDNN for the docker build
- Made ports configurable from config.json
- Add a configurable confidence threshold in config.json to filtrer out detections that are below this threshold. (default keep > 20% confidence)
- Add favicon
- Add "confirmation" modal when deleting recordings
- Add version number in side menu
- Updated all javascript dependencies: Next.js , Redux, React etc etc...
- Revert back to using the main ./darknet API as memory leak has been also fixed
v2.1.0
- Better run from file support: auto-restart, record from beginning, persist frame index #88
- Add confidence score data
- Improve reliability on desktop for long run on file
- Restore last defined counter areas on boot
- Support to input video webcam stream with "-c 0" flag without gstreamer (default darknet)