Pinned post
What if you could ๐ต๐ฆ๐ญ๐ญ vision encoders ๐ธ๐ฉ๐ข๐ต to encode?
Check out our latest work where we introduce ๐ฆ๐๐ฒ๐ฒ๐ฟ๐ฎ๐ฏ๐น๐ฒ ๐ฉ๐ถ๐๐๐ฎ๐น ๐ฅ๐ฒ๐ฝ๐ฟ๐ฒ๐๐ฒ๐ป๐๐ฎ๐๐ถ๐ผ๐ป๐.
Check out our latest work where we introduce ๐ฆ๐๐ฒ๐ฒ๐ฟ๐ฎ๐ฏ๐น๐ฒ ๐ฉ๐ถ๐๐๐ฎ๐น ๐ฅ๐ฒ๐ฝ๐ฟ๐ฒ๐๐ฒ๐ป๐๐ฎ๐๐ถ๐ผ๐ป๐.
Pretrained ViTs like DINOv2 or CLIP are great, but they produce fixed, generic representations that encode the most salient visual concepts (e.g., "cat").View quoted post on X
In human vision, prior priming with language changes how people parse an image. We believe visual encoders should do the same
๐จ Introducing Steerable Visual Representations, a new family of visual features you can steer with text towards specific visual concepts.