Tracking-Anything-with-DEVA icon indicating copy to clipboard operation
Tracking-Anything-with-DEVA copied to clipboard

How do I reproduce the results of the rvos report?

Open xinbaiw opened this issue 1 year ago • 11 comments
trafficstars

The inference is that the model being called is DEVA-propagation.pth? I used this model to infer the J&F on the rvos task, which is 64.8. The one reported in the paper is 66.0. Can you help me find out where I went wrong?

image

xinbaiw avatar Mar 24 '24 07:03 xinbaiw

The inference is that the model being called is DEVA-propagation.pth? I used this model to infer the J&F on the rvos task, which is 64.8. The one reported in the paper is 66.0. Can you help me find out where I went wrong?

image

image

xinbaiw avatar Mar 24 '24 07:03 xinbaiw

Can you compare your output with our pre-computed one and see if there is any discrepancy? It should reproduce exactly. There is too little information here for me to look into it.

hkchengrex avatar Mar 24 '24 16:03 hkchengrex

Can you compare your output with our pre-computed one and see if there is any discrepancy? It should reproduce exactly. There is too little information here for me to look into it.

Can you find anything wrong with these details?

parser.add_argument('--mask_path', default='/home/data2/Ref-YouTubeVOS-ReferFormer-SwinL-detections') parser.add_argument('--model', default='./saves/DEVA-propagation.pth') parser.add_argument('--img_path', default='/home/data2/rvos/valid/JPEGImages')

image image

xinbaiw avatar Mar 25 '24 08:03 xinbaiw

Hi, can you help me?

xinbaiw avatar Mar 26 '24 11:03 xinbaiw

I cannot peek into your file system and know what is in each folder. One potential reason is that you are not using all the frames in YouTubeVOS (i.e., https://drive.google.com/drive/folders/16uw2j3o8Pi2xhvgGopYMKyjKtzV9RQ6c)

hkchengrex avatar Mar 27 '24 01:03 hkchengrex

I cannot peek into your file system and know what is in each folder. One potential reason is that you are not using all the frames in YouTubeVOS (i.e., https://drive.google.com/drive/folders/16uw2j3o8Pi2xhvgGopYMKyjKtzV9RQ6c)

Thank you for your reply. When I used all frames of youtube-vos to test, out of memory, single card 4090 (24g), do you have a solution?

xinbaiw avatar Mar 27 '24 09:03 xinbaiw

How to reduce memory usage when testing?

xinbaiw avatar Mar 27 '24 12:03 xinbaiw

See here https://github.com/hkchengrex/Tracking-Anything-with-DEVA/blob/main/docs/EVALUATION.md#general-arguments

Use --amp

hkchengrex avatar Mar 28 '24 05:03 hkchengrex

See here https://github.com/hkchengrex/Tracking-Anything-with-DEVA/blob/main/docs/EVALUATION.md#general-arguments

Use --amp

After using amp, the memory is still not enough. Is there any other way?

longmalongma avatar Mar 29 '24 02:03 longmalongma

See here https://github.com/hkchengrex/Tracking-Anything-with-DEVA/blob/main/docs/EVALUATION.md#general-arguments Use --amp

After using amp, the memory is still not enough. Is there any other way?

Why does the test go out or memory at a later stage?

longmalongma avatar Mar 31 '24 15:03 longmalongma

推断是调用的模型是 DEVA-propagation.pth?我用这个模型推断了 rvos 任务上的 J&F,是 64.8。论文中报告的是 66.0。你能帮我找出我错在哪里吗?

图像 hi,Can you tell me how to submit the result, I have not responded after clicking submit.

wqqqqq0720 avatar Sep 05 '24 03:09 wqqqqq0720