video-result-viewer

Minimalistic video player that allows visualization, easier interpretation and analysis of synchronized metadata from FrVD dataset annotations and corresponding video files.

Description

Please refer to --help option of the tool's viewer script for all available options. See also the usage details for information about available features.

Public Use

The tool employs the following components as input:

• [required]: Original video file (production)
• [required]: Metadata annotation file from the relevant video in FrVD dataset.

Internal Use by CRIM

Following files are employed to generate metadata provided by FrVD. They are accessible only internally.

The tool takes as input the following information and displays them for corresponding timestamps.

• [required] Original video file (production)
• [optional] Extracted video-description annotations metadata (as generated by FAR-VVD/data-extraction)
• [optional] Video action inference results (as generated by FAR-VVD/video-action-recognition)
• [optional] Text subjects and verbs results (as generated by FAR-VVD/Annotation-VD)

It can also combine all of those results into a merged file (i.e.: annotations of FrVD) with aligned timestamps.

NOTE
Further options are available using the CLI script.

Changes

Please refer to Change History for any updates relative to the code or metadata adjustments.

Expected Formats of Metadata

Please refer to format details.

Installation and Execution

Please refer to usage details.

License

This work is licensed under a Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International License.

Citation

Please refer to FrVD dataset citation.