nielsr

/

vit-large-patch16-v-jepa

like 8

Hey guys! It would be useful to show how to extract video features using V-JEPA. Would you add it? You have a little snipped of code in case of the DinoV2 network (but only for images). I think it would be useful to have a code describing how to extract video (but also image) features in the case of V-JEPA :)