[Bug]

[HabibaAbderrahim]

Prerequisite

• [X] I have searched Issues and Discussions but cannot get the expected help.
• [X] The bug has not been fixed in the latest version(https://github.com/open-mmlab/mmpose).

Environment

Reproduces the problem - code sample

python demo/inferencer_demo.py tests/data/segbig/input.jpg --pose2d yoloxpose_s_8xb32-300e_coco-640.py --pose2d-weights yoloxpose_s_8xb32-300e_coco-640-56c79c1f_20230829.pth --pose3d human3d --vis-out-dir vis_results/human3d/output.jpg --device cpu

Reproduces the problem - command or script

python demo/inferencer_demo.py tests/data/segbig/input.jpg --pose2d yoloxpose_s_8xb32-300e_coco-640.py --pose2d-weights yoloxpose_s_8xb32-300e_coco-640-56c79c1f_20230829.pth --pose3d human3d --vis-out-dir vis_results/human3d/output.jpg --device cpu

Reproduces the problem - error message

ValueError: cannot select an axis to squeeze out which has size not equal to one

Additional information

I just wanted to inference 3D human pose estimation using yolox-pose as 2D keypoint detector .

[Ben-Louis]

Thank you for pointing out this problem. 3D inferencer is not compatible with one-stage methods currently. We will fix this bug soon.