Perception Encoder
Collection
OpenCLIP (PE Core image + text) and timm PE Core, Spatial, Lang (ViT only) weights. NOTE: These weights do not work with original modeling code.
โข
19 items
โข
Updated
โข
7