When input is a video, only the previous 37 frames are processed and saved.

MVIG-SJTU / AlphaPose

Real-Time and Accurate Full-Body Multi-Person Pose Estimation&Tracking System

http://mvig.org/research/alphapose.html

Other

7.96k stars 1.97k forks source link

When input is a video, only the previous 37 frames are processed and saved. #927

Closed Huhaowen0130 closed 2 years ago

Huhaowen0130 commented 2 years ago

Hello, when I ran "python scripts/demo_inference.py --cfg configs/coco/resnet/256x192_res50_lr1e-3_1x-simple.yaml --checkpoint pretrained_models/simple_res50_256x192.pth --video examples/GH010011.MP4 --outdir examples /res --save_video" on Ubuntu, only previous 37 frames were processed. Could you help me whith it? Thank you!

Fang-Haoshu commented 2 years ago

It denotes your video can be only decoded into 37 frames by opencv. Perhaps you can try some other format or video sources.