Add video inference

[sokovninn]

This pull request includes several changes to enhance the inference capabilities of the luxonis_train module, particularly by adding support for video files and improving the handling of image and video sources. The most important changes include renaming variables for clarity, adding support for video processing, and updating the inference logic to accommodate these improvements.

Enhancements to Inference Capabilities:

• luxonis_train/__main__.py: Renamed ImgPathType to SourcePathType and updated the help text to include video files. Updated the infer function to use source_path instead of img_path.

• luxonis_train/core/core.py: Added process_video to the imports. Defined IMAGE_FORMATS and VIDEO_FORMATS sets for supported file types. Updated the infer method to handle video files and directories containing images.

Improvements in Utility Functions:

• luxonis_train/core/utils/infer_utils.py: Added process_video function to handle video file inference. Refactored process_images and introduced prepare_and_infer_image for better code reuse. Updated render_visualizations to optionally return visualizations instead of displaying them.

Possible improvements:

• Parallel video processing and batched inference
• Webcam/stream source for inference

[klemen1999]

LGTM, left some comments

[codecov]

Codecov Report

Attention: Patch coverage is 20.00000% with 44 lines in your changes missing coverage. Please review.

Please upload report for BASE (dev@c97f921). Learn more about missing BASE report.

✅ All tests successful. No failed tests found.

Files with missing lines	Patch %	Lines
luxonis_train/core/utils/infer_utils.py	21.27%	37 Missing ⚠️
luxonis_train/core/core.py	12.50%	7 Missing ⚠️

Additional details and impacted files

@@          Coverage Diff           @@
##             dev      #84   +/-   ##
======================================
  Coverage       ?   96.04%           
======================================
  Files          ?      143           
  Lines          ?     6341           
  Branches       ?        0           
======================================
  Hits           ?     6090           
  Misses         ?      251           
  Partials       ?        0           

☔ View full report in Codecov by Sentry.
📢 Have feedback on the report? Share it here.

[kozlov721]

LGTM