Alexandre Morgand (@Almorgand)

2025-05-12 | โค๏ธ 182 | ๐Ÿ” 22


DiffusionSfM: Predicting Structure and Motion via Ray Origin and Endpoint Diffusion

TL;DR: pixel-wise ray origins and endpoints in a global frame; denoising diffusion process; patch-wise embeddings with DINOv2 and embed noisy ray origins and endpoints into latents (1/2) https://x.com/Almorgand/status/1921852832149295413/video/1

๐Ÿ”— ์›๋ณธ ๋งํฌ

๋ฏธ๋””์–ด

video


Auto-generated - needs manual review

Tags

domain-vision-3d domain-genai domain-visionos