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.

Sign up for GitHub

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

Jump to bottom

Improve video analyzer #68

Closed
5 tasks done
mamaheux opened this issue Dec 15, 2022 · 0 comments · Fixed by #74
Closed
5 tasks done

Improve video analyzer #68

mamaheux opened this issue Dec 15, 2022 · 0 comments · Fixed by #74

Comments

@mamaheux
Copy link
Collaborator

mamaheux commented Dec 15, 2022 •
edited
Loading

  • Object Detection
    • Add the class/box confidence value to the message
  • Face descriptor
    • Add a value measuring the blur of the face to the message (std of laplacian filter output)
    • Add a flag indicating how the face is aligned to the message
@mamaheux mamaheux changed the title Improve video analyzer face descriptor Improve video analyzer Apr 27, 2023
mamaheux added a commit that referenced this issue Apr 27, 2023
@mamaheux
Improve video analyzer (ref #68)
1eaf3e4
@mamaheux mamaheux linked a pull request Apr 28, 2023 that will close this issue
Improve dnn #74
Merged
mamaheux added a commit that referenced this issue Nov 15, 2023
@mamaheux
Improve dnn (#74)
ae7d1f9 
* Update pose_estimor and face_descriptor_extractor

* Fix heatmap_sigma type

* Improve face alignment, fix face_descriptor extractor training.

* Improve audio analyzer (ref #70).

* Improve video analyzer (ref #68)

* Update HBBA lite.

* Update video analysis and audio analysis loggers.

* Add HBBA strategies log to the perception logger.

* Rename blur score to sharpness score.

* Add distillation_loss_alpha to train_pose_estimator.py

* Add distillation_loss_alpha to output path.

* Fix trainers for criterion next_epoch. Fix export_yolo_v4. Add GPU support to align_faces.py and test_pose_estimator_with_yolo_v4.py.

* Fix OpenFaceBackbone

* Add yolov7 module.

* Reformat code

* Add yolov7 module.

* Add missing eval in test_converted_yolo.

* Fix YOLO V7. Add the choice of the yolo model in video_analyzer.

* Add AudioSet support. Add MS-celeb-support.

* Add margin to the output path.

* Fix dnns

* Set the odas frame in the midlle of the mics.

* Use the face_sharpness_score in person_identification. Fix test launch files.

* Update video_analyzer dnns export.

* Add a model choice for FaceDescriptorExtractor.

* Fix dnn_units tests.

* Fix the realsense in capture_face.launch

* Add PSLA pooling. Add backends.

* Add mixup and VIT to backbone training.

* Add missing changes for the backbone training.

* Add passt_s_n models to the audio trainings.

* Use the old networks for the face description.

* Refactoring for descriptor_yolo_v7 and yolov7 trained on Objects365.

* Fix test_face_descriptor_extractor

* Fix descriptor Yolov7

* Fix descriptor yolov7

* Fix YOLOV7 trained on Objects365.

* Fix video analyzer and export.

* Fix issues.

* Fix dnn_utils tests.

* Fix dnn_utils tests.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging a pull request may close this issue.

Improve dnn introlab/t-top
1 participant
@mamaheux