MeshRipple, a new framework that generates high-face-count, artist-quality 3D meshes via an autoregressive approach, has been released and accepted as a Highlight at CVPR 2026, a leading computer vision conference. The paper was posted to arXiv on December 8, 2025, with inference code and checkpoints available on a project page and GitHub. The authors are led by Junkai Lin and Hang Long of Huazhong University of Science and Technology, with Matthias Nießner of TUM among the co-authors.
Continue reading
The rest of this article is for AI News Blitz readers. Choose an option below to keep reading.
Already purchased? Sign in✓ Signed in — this article isn’t included in your current plan.