Kwang Moo Yi (@kwangmoo_yi)
2025-07-16 | ❤️ 108 | 🔁 24
Preprint of today: Zhuo and Zheng et al., “Streaming 4D Visual Geometry Transformer” — https://wzzheng.net/StreamVGGT/
VGGT with cache / causal attention for 70ms inference on an image stream. Similar to other Dust3R speed-up methods, but with VGGT. https://x.com/kwangmoo_yi/status/1945528288044355983/photo/1
🔗 Related
See similar notes in domain-vision-3d, domain-llm