Inference on own dataset frames?

[antithing]

Hi., and thank you for this code!

Can you please let me know the command to run inference on my own dataset (image sequences), using IDOL and your pre-trained model?

Thank you!

[antithing]

I have a demo py script, and am running it on a video file, however I get the following error:

[07/07 10:23:03 fvcore.common.checkpoint]: [Checkpointer] Loading from models/OVIS_SwinL_414AP.pth ...
  0%|                                                                                          | 0/161 [00:00<?, ?it/s]d:\matting\vnext\projects\IDOL\idol\models\tracker.py:213: UserWarning: To copy construct from a tensor, it is recommended to use sourceTensor.clone().detach() or sourceTensor.clone().detach().requires_grad_(True), rather than torch.tensor(sourceTensor).
  mask_new_indices = torch.tensor(indices)[valids].tolist()
  0%|                                                                                          | 0/161 [00:01<?, ?it/s]
Traceback (most recent call last):
  File "D:\Matting\VNext\projects\IDOL\demo_idol.py", line 176, in <module>
    for vis_frame in tqdm.tqdm(demo.run_on_video(video) , total=num_frames ):
  File "C:\Users\B\AppData\Local\Programs\Python\Python39\lib\site-packages\tqdm\std.py", line 1195, in __iter__
    for obj in iterable:
  File "d:\matting\vnext\demo\predictor.py", line 129, in run_on_video
    yield process_predictions(frame, self.predictor(frame))
  File "d:\matting\vnext\detectron2\engine\defaults.py", line 317, in __call__
    predictions = self.model([inputs])[0]
KeyError: 0

What might be happening here?

[antithing]

Ah, using this repo:

https://github.com/reno77/VNext

I have it working.