-
Notifications
You must be signed in to change notification settings - Fork 95
New issue
Have a question about this project? # for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “#”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? # to your account
RL--based Learner for Active Face Recognition #473
Conversation
…n/tester.py removed extra file
…learner_fr_and_pose
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you @Pavlos-Tosidis for the new tool! I added several comments about some fixes. One general comment that i also noted in the readme doc, is that it would be helpful to add a demo.py file that runs a simple evaluation (or similar) loop, so a user can run this out-of-the-box following the documentation and seeing a trained agent in action, without the need to create their own demo file.
...tion/active_perception/active_face_recognition/simulation/worlds/active_face_recognition.wbt
Show resolved
Hide resolved
...endr/perception/active_perception/active_face_recognition/active_face_recognition_learner.py
Show resolved
Hide resolved
...endr/perception/active_perception/active_face_recognition/active_face_recognition_learner.py
Outdated
Show resolved
Hide resolved
...endr/perception/active_perception/active_face_recognition/active_face_recognition_learner.py
Outdated
Show resolved
Hide resolved
Ignored num_steps for eval() but fixed num_episodes logic. Implemented suggestions on ReadME
Co-authored-by: Kostas Tsampazis <27914645+tsampazk@users.noreply.github.com>
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you for this great contribution. That looks very good to me. Any reason why you are not using the latest version of Webots, e.g., R2023b?
minor fixes on README
…learner_fr_and_pose
minor fixes on README
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thank you!
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Very good. Thank you.
* Active_fr * Active_fr * Active_fr * Fixes and ReadME * Fixes * Fixes and ReadME * Delete src/opendr/perception/active_perception/active_face_recognition/tester.py removed extra file * Fixes and ReadME * Attempted merge fix for continual * Removed some prints * Face images for database creation * Eval fix * Fixed suggestions by Kostas. Ignored num_steps for eval() but fixed num_episodes logic. Implemented suggestions on ReadME * Added demo under projects * Added demo under projects and corresponding README * Blank lines on demos * Blank lines on demos * Update docs/reference/active_face_recognition.md Co-authored-by: Kostas Tsampazis <27914645+tsampazk@users.noreply.github.com> * Fixed Webots version on protos files, minor fixes on README * Fixed Webots version on protos files, minor fixes on README --------- Co-authored-by: Kostas Tsampazis <27914645+tsampazk@users.noreply.github.com>
Adding Active Face Recognition agent. The implementation relies on stable_baselines3 and trains on Webots.
A Webots world is provided and the corresponding environment file to train an agent in the task of Active Face Recognition, in which the goal is to maximize FR confidence while controlling a Mavic2 drone