Thank you for your work and for the detailed documentation. I am running AlphaPose on a video and cannot get it to predict keypoints for every frame. I have changed the confidence parameter to 0 in AlphaPose/configs/coco/resnet/256x192_res50_lr1e-3_1x.yaml, but am still not able to get pose predictions for every frame.
And here is the command I am using: !python3 scripts/demo_inference.py --cfg configs/coco/resnet/256x192_res50_lr1e-3_1x.yaml --checkpoint pretrained_models/fast_res50_256x192.pth --video /content/video.mp4 --save_video --outdir /content/output_alphapose
Hello,
Thank you for your work and for the detailed documentation. I am running AlphaPose on a video and cannot get it to predict keypoints for every frame. I have changed the confidence parameter to 0 in
AlphaPose/configs/coco/resnet/256x192_res50_lr1e-3_1x.yaml
, but am still not able to get pose predictions for every frame.Here is the video (and resulting video/keypoints) I am performing inference on: https://drive.google.com/drive/folders/1MC5gkDl7M6vWRbbho8o50H5_KtgUSnZ4?usp=sharing
And here is the command I am using:
!python3 scripts/demo_inference.py --cfg configs/coco/resnet/256x192_res50_lr1e-3_1x.yaml --checkpoint pretrained_models/fast_res50_256x192.pth --video /content/video.mp4 --save_video --outdir /content/output_alphapose
Any suggestions for solving this?
Thanks!