facebookresearch / omnivore

Omnivore: A Single Model for Many Visual Modalities
Other
559 stars 38 forks source link

How to process video data when the input is RGB-D image. #46

Open Zhangwenyao1 opened 5 months ago

Zhangwenyao1 commented 5 months ago

Thanks for your great work! I'm interested in understanding the procedure for handling video data during training if the input comprises RGB-D images. Do you simply set them to zero, or is there another approach?