search

Epiphqny / VisTR

[CVPR2021 Oral] End-to-End Video Instance Segmentation with Transformers
https://arxiv.org/abs/2011.14503
Apache License 2.0
738 stars 96 forks source link

python inference.py --masks --model_path /home/zhangsai/vistr/vistr_r50.pth #61

Open qzsrh opened 2 years ago

qzsrh commented 2 years ago

When I run this command, :python inference.py --masks --model_path /home/zhangsai/vistr/vistr_r50.pth the following error occurs： Process video: 292 Process video: 293 Traceback (most recent call last): File "inference.py", line 236, in main(args) File "inference.py", line 217, in main if pred_scores[n,m]<0.001: IndexError: index 36 is out of bounds for axis 0 with size 36 My dataset is YoutubeVIS，The data format is the same as that of preparation. The Val dataset is used to execute this line of command. Hope to get your reply, thank you very much!

XiangTodayEatsWhat commented 2 years ago

Hello! Have you addressed it? I met the same problem.

qzsrh commented 2 years ago

Hello! Have you addressed it? I met the same problem.

Some of the videos in the test set exceed 36 frames. Just find them and remove them

XiangTodayEatsWhat commented 2 years ago

Hello! Have you addressed it? I met the same problem.

Some of the videos in the test set exceed 36 frames. Just find them and remove them

Thank you! It works now!

qzsrh commented 2 years ago

Hello! Have you addressed it? I met the same problem.

Some of the videos in the test set exceed 36 frames. Just find them and remove them

Thank you! It works now!

However, it seems that there are errors in the json file after my test. I didn't successfully visualize it. If you can successfully visualize it later, please tell me what you did. Thank you very much!

shuangkouyizu commented 1 year ago

Have you solved it？