search

MVIG-SJTU / AlphaPose

Real-Time and Accurate Full-Body Multi-Person Pose Estimation&Tracking System
http://mvig.org/research/alphapose.html
Other
8.06k stars 1.98k forks source link

遇到Segmentation fault #119

Closed YogaHe closed 6 years ago

YogaHe commented 6 years ago

执行视频命令时，Segmentation fault

Jeff-sjtu commented 6 years ago

请使用此仓库安装torchsample

ghost commented 6 years ago

Hello, "No module named 'torchsample'" is this error fixed ? please provide the fix.

Jeff-sjtu commented 6 years ago

Please use this repo torchsample. python setup --install

guancheng817 commented 5 years ago

I met same problem,Segmentation fault (core dumped),then i install the repo torchsample, but the issue still appeared, can u help me ?

Jeff-sjtu commented 5 years ago

Please give me the whole error info. @guancheng817

guancheng817 commented 5 years ago

(torch0.4) ouc@ouc-titanx:/media/ouc/4T_B/gc/AlphaPose$ python3 video_demo.py --video ./video/Panama.mp4 --outdir examples/res --save_video Segmentation fault (core dumped) (torch0.4) ouc@ouc-titanx:/media/ouc/4T_B/gc/AlphaPose$

it's the whole error info. Thanks.. @Jeff-sjtu

guancheng817 commented 5 years ago

Please give me the whole error info. @guancheng817

The whole error info only one line ..

guancheng817 commented 5 years ago

I recreate a conda enviroment, and everything is ok...

dmg206 commented 5 years ago

I have met this problem too. It didn't OK for the torchsampe. can u help me.the log is below:

% /home/em/anaconda3/envs/alphapose_pytorch/bin/python3 demo.py --indir examples/demo --outdir examples/res/ --vis --save_img Loading YOLO model.. Loading pose model from ./models/sppe/duc_se.pth 100%|███████████████████████████████████████████████████████████████████████████████████| 1/1 [00:00<00:00, 4.43it/s] ===========================> Finish Model Running. ===========================> Rendering remaining images in the queue... ===========================> If this step takes too long, you can enable the --vis_fast flag to use fast rendering (real-time). zsh: segmentation fault (core dumped) /home/em/anaconda3/envs/alphapose_pytorch/bin/python3 demo.py --indir --vi

wine3603 commented 5 years ago

我安装的是pytorch-0.4 版本，也是同样的问题。。。 ~/AlphaPose$ python3 demo.py --indir examples/demo --outdir examples/res --save_img --vis_fast Loading YOLO model.. Loading pose model from ./models/sppe/duc_se.pth 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3/3 [00:00<00:00, 7.09it/s] ===========================> Finish Model Running. Segmentation fault (core dumped) or 'python3 demo.py --indir examples/demo/ --outdir examples/res --save_img --vis_fast Loading YOLO model.. Loading pose model from ./models/sppe/duc_se.pth 100%|█████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████████| 3/3 [00:00<00:00, 7.67it/s] ===========================> Finish Model Running. Segmentation fault (core dumped)'

wine3603 commented 5 years ago

I solved this bug by down grade the pytorch = 0.4.0! my former version was 0.4.1........ it must be pytorch=0.4.0, other version meet Segmentation fault (core dumped)'

carolynenewman commented 5 years ago

I have downgraded to pytorch=0.4.0, and have torchsample installed. I also have one .jpg image in a directory which is accurately called upon in the command line. Yet, I am still getting a segmentation fault. Has anyone found another solution to this problem?

allenday commented 4 years ago

pip install torch==0.4.0 torchvision==0.2.1 Pillow==6.1