Hierarchical Neural Surfaces for 3D Mesh Compression

Sai Karthikey Pentapati, Gregoire Phillips, Alan C. Bovik · 2025

Read Hierarchical Neural Surfaces for 3D Mesh Compression by Sai Karthikey Pentapati, Gregoire Phillips, Alan C. Bovik (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2512-15985.txt.

Description

Sai Karthikey Pentapati, Gregoire Phillips, and Alan Bovik present hierarchical neural surfaces for 3D mesh compression in a 2025 paper. Implicit neural representations compress images, video, and audio well, but most work encodes signed distance functions rather than the triangle meshes that industry actually uses. Bridging that gap is the aim here.

Their method maps a zero genus mesh surface onto a unit sphere, then encodes the displacement field that rebuilds the original shape as a neural representation. A hierarchical design recovers coarse structure first and fine detail later, which keeps the code compact. Meshes of arbitrary resolution can be decoded in real time, with state of the art quality for the compressed size.

Catalog record

CREATORSai Karthikey Pentapati, Gregoire Phillips, Alan C. Bovik
YEAR2025
COLLECTIONPAPERS
FORMATTXT
FILE SIZE22 KB
VERSION1.0
LICENSEhttp://creativecommons.org/licenses/by/4.0/
RIGHTSCC BY
SOURCEarXiv

Source: https://arxiv.org/abs/2512-15985

Browse the full library · About the Tower of Babel