Georgios Pavlakos (@geopavlakos)

2025-05-05 | โค๏ธ 49 | ๐Ÿ” 5


Make sure to check out Hanwenโ€™s @hanwenjiang1 latest work! ๐Ÿš€ We introduce RayZer, a self-supervised model for novel view synthesis. We use zero 3D supervision, yet we outperform supervised methods! Some surprising and exciting results inside! ๐Ÿ”๐Ÿ”ฅ


Auto-generated - needs manual review

์ธ์šฉ ํŠธ์œ—

Hanwen Jiang (@hanwenjiang1)

Supervised learning has held 3D Vision back for too long.

Meet RayZer โ€” a self-supervised 3D model trained with zero 3D labels: โŒ No supervision of camera & geometry โœ… Just RGB images

And the wild part? RayZer outperforms supervised methods (as 3D labels from COLMAP is noisy)

๐ŸŒ Project: https://t.co/YpHJIu0tfs (1/4)

์›๋ณธ ํŠธ์œ—

๐ŸŽฌ ์˜์ƒ

Tags

domain-vision-3d domain-ai-ml