Andrei Bursuc (@abursuc)

2024-10-25 | โค๏ธ 153 | ๐Ÿ” 18


If you liked UNIC, check out Theia: a simple approach for distilling pretrained models, e.g.,CLIP, DINOv2, SAM, DepthAnything into a unified model with improved performance on policy learning https://theia.theaiinstitute.com/ https://x.com/abursuc/status/1849945966536950080/video/1

๋ฏธ๋””์–ด

video

photo

photo

photo

์ธ์šฉ

[@abursuc]: Excellent work by @mbsariyildiz et al. on distilling multiple complementary visual encoders into a single one. This is particularly useful in the era of pretrained foundation models on different dataโ€ฆ


์ธ์šฉ ํŠธ์œ—

Andrei Bursuc (@abursuc)

Excellent work by @mbsariyildiz et al. on distilling multiple complementary visual encoders into a single one. This is particularly useful in the era of pretrained foundation models on different datasets and with different types of supervision ๐Ÿ‘‡

์›๋ณธ ํŠธ์œ—

Tags

domain-reconstruction