vita-epfl / DePOSit

[ICRA 2023] Official implementation of "A generic diffusion-based approach for 3D human pose prediction in the wild".
GNU Affero General Public License v3.0
49 stars 14 forks source link

Regarding Usability as 3D Human Pose Keypoint Repair Pipeline #6

Open aviralchharia opened 7 months ago

aviralchharia commented 7 months ago

Hi authors, Great work! The study mentions that post-processing can be used to enhance the human pose of any prediction model. Can the authors comment if it can be used for repairing key points. For example, if we have only 14 predicted key points in COCO format (from any model like MediaPipe/ YOLOv8), and 3 missing key points. Can the approach be used to predict the remaining key points? Thanks!

satijapratik commented 5 months ago

I have the same question^. Can this be addressed?

SaeedSaadatnejad commented 5 months ago

Hi, yes it can be used for repairing too. You just need to set the appropriate observation length and apply TCD on your sequences.