Inference on my own video clips

[heldJan]

Hello,

I was wondering if it is possible to perform inference on my own video?

Thank you!

[Tortoise17]

@SilvioGiancola is there any inference pipeline available to test the model on customized video?

[SilvioGiancola]

Yes, you can run inference on a custom dataset containing your own video with this code: https://github.com/OpenSportsLab/OSL-ActionSpotting/blob/main/tools/infer.py

[Tortoise17]

@SilvioGiancola Just a quick question, can you guide me, dataset.test.path=/datasets/SoccerNet/ResNET_PCA512/test/annotations.json looking into the part of inference command, how to create annotations.json for the custom video? From which tool we can create for using this repo. I mean not manually.

[Tortoise17]

@SilvioGiancola Basically, what I am trying to do is, like in your soccernet_dev_kit, there is CALF base architecture, and for that inference workflow is available which can extract fratures and using those, it generates the file which is equal to Labels_v2.json. and with this infer.py file, I want to extract the labels_v2.json using the netvlad++ netrvlad++. Still unsuccessful. If you can guide or hint.

[Tortoise17]

@SilvioGiancola I tried the file generated by CALF inference from your working SoccerNet-Dev-Kit, named as annotations, for this but still the error is :

command line is
(football) [u120332@machine1 OSL-ActionSpotting]$ python tools/infer.py python tools/infer.py --cfg-options configs/learnablepooling/json_netvlad++_resnetpca512.py dataset.test.data_root=/scratch/local1/tmp/OSL-ActionSpotting/data/input/soccer_test dataset.test.path=/scratch/local1/tmp/OSL-ActionSpotting/annotations/annotations.json

Generated resulting error is

Traceback (most recent call last):
  File "/scratch/local1/tmp/OSL-ActionSpotting/tools/infer.py", line 161, in <module>
    main()
  File "/scratch/local1/tmp/OSL-ActionSpotting/tools/infer.py", line 43, in main
    args = parse_args()
  File "/scratch/local1/tmp/OSL-ActionSpotting/tools/infer.py", line 37, in parse_args
    args = parser.parse_args()
  File "/home/python/anaconda3/envs/football/lib/python3.10/argparse.py", line 1833, in parse_args
    args, argv = self.parse_known_args(args, namespace)
  File "/home/python/anaconda3/envs/football/lib/python3.10/argparse.py", line 1866, in parse_known_args
    namespace, args = self._parse_known_args(args, namespace)
  File "/home/python/anaconda3/envs/football/lib/python3.10/argparse.py", line 2079, in _parse_known_args
    start_index = consume_optional(start_index)
  File "/home/python/anaconda3/envs/football/lib/python3.10/argparse.py", line 2019, in consume_optional
    take_action(action, args, option_string)
  File "/home/python/anaconda3/envs/football/lib/python3.10/argparse.py", line 1943, in take_action
    action(self, namespace, argument_values, option_string)
  File "/home/python/anaconda3/envs/football/lib/python3.10/site-packages/mmengine/config/config.py", line 1840, in __call__
    key, val = kv.split('=', maxsplit=1)
ValueError: not enough values to unpack (expected 2, got 1)

Please just guide if I am missing some point.