Google SIGGRAPH paper rebuilds 3D heads in 0.08 seconds with 88% less GPU memory
rsasaki0109 · x · 2026-07-27
Google’s SIGGRAPH 2026 paper, “Topologically Consistent Multi-view 3D Head Reconstruction via Coarse-Guided Layered Surface Sampling,” introduces SHELLS, a method that reconstructs dense 3D heads from calibrated multi-view images.
Key results in the post:
- reconstructs 18k-vertex 3D heads in 0.08 seconds
- uses DINOv2 features aggregated through projective surface-aware feature sampling
- lets a transformer predict dense semantic meshes
- is 3.5× faster than state of the art
- uses 88% less GPU memory
The main claim is that this produces an efficient reconstruction pipeline with better global consistency across views.
More from Multimodal
- AI-Generated Cat Adventure Videos Go Viral with Over 20M Views — aziz4ai · 2026-07-28
- Hugging Face open-sources a local speech-to-speech voice-agent stack — huggingface · 2026-07-28
- AI Rebuilds Homer's Odyssey into a 135-Minute Feature Film — heypearlai · 2026-07-28
- An AI remake of The Odyssey trailer scenes is weird enough to be worth sharing — eyishazyer · 2026-07-28
- Nvidia puts an open vision-language-action model on Hugging Face — theteknosaur · 2026-07-28
- A Grok user asks to turn Hodor into Shrek in a classic AI meme prompt — heypearlai · 2026-07-28