OpenMMLab Video Perception Toolbox. It supports Video Object Detection (VID), Multiple Object Tracking (MOT), Single Object Tracking (SOT), Video Instance Segmentation (VIS) with a unified framework.
I've encountered an issue where the JSON files generated by imagenet2coco_vid.py are significantly larger than what was provided in the TransVOD documentation. I am curious if there's a particular method or setting within the script that applies data compression, which might not be immediately obvious from the code or the documentation.
Could you please provide some guidance on how to reduce the size of the output JSON files? Is there a specific option or technique that could be used to achieve a more compact representation?
Thank you very much for your attention to this matter and any assistance you can offer.
I've encountered an issue where the JSON files generated by
imagenet2coco_vid.pyare significantly larger than what was provided in the TransVOD documentation. I am curious if there's a particular method or setting within the script that applies data compression, which might not be immediately obvious from the code or the documentation.Could you please provide some guidance on how to reduce the size of the output JSON files? Is there a specific option or technique that could be used to achieve a more compact representation?
Thank you very much for your attention to this matter and any assistance you can offer.