demo的video无法跑出结果 about yolox HOT 17 CLOSED

Don't use yolox_ s. Pth.tar, author's yolox.py didn't have get it_ Model () method,So the weight file cannot be loaded

from yolox.

Don't use yolox_ s. Pth.tar, author's yolox.py didn't have get it_ Model () method,So the weight file cannot be loaded

But I can run image successfully?

from yolox.

[mov,mp4,m4a,3gp,3g2,mj2 @ 0x558102961000] moov atom not found
VIDIOC_REQBUFS: Inappropriate ioctl for device

from yolox.

it seems complain about video file format??

from yolox.

似乎抱怨视频文件格式？

估计是，那应该用什么视频格式呢

from yolox.

source code using either real time camera or video file.
I looked up the error message that it is meta info about video file.

from yolox.

I tried another mp4 file, it works. so It is about video file compatibility.

from yolox.

我尝试了另一个 mp4 文件，它有效。所以这是关于视频文件的兼容性。

I tried another mp4 file, it works. so It is about video file compatibility.

I use MP4 format about 280M. Because it's too big？Or could u show your operation?

from yolox.

no. similar to image file, there is meta info about whole video file. some have, some don't.
I downloaded another mp4 file, it works. same operation.

from yolox.

!python tools/demo.py video -n yolox-x -c /content/YOLOX/yolox_x.pth.tar --path /content/YOLOX/airport.mp4 --conf 0.3 --nms 0.65 --tsize 640 --save_result

2021-07-22 07:51:38 | INFO | main:219 - Args: Namespace(camid=0, ckpt='/content/YOLOX/yolox_x.pth.tar', conf=0.3, demo='video', exp_file=None, experiment_name='yolox_x', fp16=False, fuse=False, name='yolox-x', nms=0.65, path='/content/YOLOX/airport.mp4', save_result=True, trt=False, tsize=640)
/usr/local/lib/python3.7/dist-packages/torch/nn/functional.py:718: UserWarning: Named tensors and all their associated APIs are an experimental feature and subject to change. Please do not use them for anything important until they are released as stable. (Triggered internally at /pytorch/c10/core/TensorImpl.h:1156.)
return torch.max_pool2d(input, kernel_size, stride, padding, dilation, ceil_mode)
2021-07-22 07:51:39 | INFO | main:229 - Model Summary: Params: 99.07M, Gflops: 281.93
2021-07-22 07:51:43 | INFO | main:240 - loading checkpoint
2021-07-22 07:51:44 | INFO | main:245 - loaded checkpoint done.
2021-07-22 07:51:44 | INFO | main:183 - video save_path is ./YOLOX_outputs/yolox_x/vis_res/2021_07_22_07_51_44/airport.mp4
2021-07-22 07:51:46 | INFO | main:128 - Infer time: 1.9449s
2021-07-22 07:51:46 | INFO | main:128 - Infer time: 0.0612s
2021-07-22 07:51:46 | INFO | main:128 - Infer time: 0.0587s
2021-07-22 07:51:46 | INFO | main:128 - Infer time: 0.0586s
2021-07-22 07:51:46 | INFO | main:128 - Infer time: 0.0597s

from yolox.

!python tools/demo.py video -n yolox-x -c /content/YOLOX/yolox_x.pth.tar --path /content/YOLOX/airport.mp4 --conf 0.3 --nms 0.65 --tsize 640 --save_result

2021-07-22 07:51:38 | INFO | main:219 - Args: Namespace(camid=0, ckpt='/content/YOLOX/yolox_x.pth.tar', conf=0.3, demo='video', exp_file=None, experiment_name='yolox_x', fp16=False, fuse=False, name='yolox-x', nms=0.65, path='/content/YOLOX/airport.mp4', save_result=True, trt=False, tsize=640)
/usr/local/lib/python3.7/dist-packages/torch/nn/functional.py:718: UserWarning: Named tensors and all their associated APIs are an experimental feature and subject to change. Please do not use them for anything important until they are released as stable. (Triggered internally at /pytorch/c10/core/TensorImpl.h:1156.)
return torch.max_pool2d(input, kernel_size, stride, padding, dilation, ceil_mode)
2021-07-22 07:51:39 | INFO | main:229 - Model Summary: Params: 99.07M, Gflops: 281.93
2021-07-22 07:51:43 | INFO | main:240 - loading checkpoint
2021-07-22 07:51:44 | INFO | main:245 - loaded checkpoint done.
2021-07-22 07:51:44 | INFO | main:183 - video save_path is ./YOLOX_outputs/yolox_x/vis_res/2021_07_22_07_51_44/airport.mp4
2021-07-22 07:51:46 | INFO | main:128 - Infer time: 1.9449s
2021-07-22 07:51:46 | INFO | main:128 - Infer time: 0.0612s
2021-07-22 07:51:46 | INFO | main:128 - Infer time: 0.0587s
2021-07-22 07:51:46 | INFO | main:128 - Infer time: 0.0586s
2021-07-22 07:51:46 | INFO | main:128 - Infer time: 0.05
I use YOLOX-s. Is it some problem with model?

from yolox.

!python tools/demo.py video -n yolox-x -c /content/YOLOX/yolox_x.pth.tar --path /content/YOLOX/airport.mp4 --conf 0.3 --nms 0.65 --tsize 640 --save_result

2021-07-22 07:51:38 | INFO | main:219 - Args: Namespace(camid=0, ckpt='/content/YOLOX/yolox_x.pth.tar', conf=0.3, demo='video', exp_file=None, experiment_name='yolox_x', fp16=False, fuse=False, name='yolox-x', nms=0.65, path='/content/YOLOX/airport.mp4', save_result=True, trt=False, tsize=640)
/usr/local/lib/python3.7/dist-packages/torch/nn/functional.py:718: UserWarning: Named tensors and all their associated APIs are an experimental feature and subject to change. Please do not use them for anything important until they are released as stable. (Triggered internally at /pytorch/c10/core/TensorImpl.h:1156.)
return torch.max_pool2d(input, kernel_size, stride, padding, dilation, ceil_mode)
2021-07-22 07:51:39 | INFO | main:229 - Model Summary: Params: 99.07M, Gflops: 281.93
2021-07-22 07:51:43 | INFO | main:240 - loading checkpoint
2021-07-22 07:51:44 | INFO | main:245 - loaded checkpoint done.
2021-07-22 07:51:44 | INFO | main:183 - video save_path is ./YOLOX_outputs/yolox_x/vis_res/2021_07_22_07_51_44/airport.mp4
2021-07-22 07:51:46 | INFO | main:128 - Infer time: 1.9449s
2021-07-22 07:51:46 | INFO | main:128 - Infer time: 0.0612s
2021-07-22 07:51:46 | INFO | main:128 - Infer time: 0.0587s
2021-07-22 07:51:46 | INFO | main:128 - Infer time: 0.0586s
2021-07-22 07:51:46 | INFO | main:128 - Infer time: 0.0597s

I tried several mp4 videos. But it couldn't work. Could u send me your video to try?

from yolox.

I don't know what happened. It works

from yolox.

我也是换了一个mp4后可以了。

from yolox.

我有两个摄像头，其中一个录的mp4都不行，另一个都行。。

from yolox.

这是什么问题呀 解决不了 我也遇到了

from yolox.

我写在这个博客里面了 可以查看
https://blog.csdn.net/qq_43190806/article/details/122362243

或者有一个答复
tools/demo.py中的第219行，有这样一句，
save_path = os.path.join(save_folder, args.path.split("/")[-1])
当args.path不是"/"分割的时候，就会出错误，导致源文件被覆盖。
应该改成这样：
save_path = os.path.join(save_folder, args.path[len(os.path.dirname(args.path))+1:])

from yolox.