search

Walter0807 / MotionBERT

[ICCV 2023] PyTorch Implementation of "MotionBERT: A Unified Perspective on Learning Human Motion Representations"
Apache License 2.0
1.02k stars 123 forks source link

Input keypoint structure #44

Closed jlnk03 closed 1 year ago

jlnk03 commented 1 year ago

Hi, the documentation says to use the H36M keypoint format or the Halpe 26 keypoints. Since these two formats differ and I’m trying to use YOLOv7 to extract the 2D poses, which keypoints and ordering does MotionBERT expect? Is there an example json available? Thank you😊

XALEX-123 commented 1 year ago

Each input format should be alright, because you can map any keypoint into the other one in the data/dataset_wild.py

Walter0807 commented 1 year ago

Thanks @XALEX-123. Here is an example json file in AlphaPose output format. sample.zip

jlnk03 commented 1 year ago

Thanks to both of you! This helped a lot :)