---
title: "Google's CUBE Beats Old 3D Face Models With 0.95mm Precision"
subtitle: "Google researchers introduce CUBE, a hybrid B-spline and neural representation for 3D faces that beats prior scan registration methods while enabling local editing."
url: "https://alphasignal.ai/news/google-s-cube-beats-old-3d-face-models-with-0-95mm-precision"
author: "AlphaSignal Newsroom"
author_url: "https://alphasignal.ai/editorial-team"
company: "AlphaSignal"
company_url: "https://alphasignal.ai/company/alphasignal"
published: "2026-04-14T00:00:00.000Z"
updated: "2026-08-23T21:57:38.949Z"
type: "Paper"
topics:
  - "Image"
  - "Development"
subtopics:
  - "Neural Rendering"
  - "Vision Language"
upvotes: 2097
source_url: "https://arxiv.org/abs/2604.12894"
comments: 0
truncated: true
---

# Google's CUBE Beats Old 3D Face Models With 0.95mm Precision

> Google researchers introduce CUBE, a hybrid B-spline and neural representation for 3D faces that beats prior scan registration methods while enabling local editing.

A team at Google has proposed a new way to represent 3D human faces that borrows a decades-old idea from computer-aided design and gives it a neural upgrade. The representation, called [CUBE (Control-based Unified B-spline Encoding)](https://arxiv.org/abs/2604.12894), replaces the low-dimensional 3D control points of a traditional B-spline volume with high-dimensional learnable feature vectors, then bolts on a tiny MLP to add fine geometric detail. The result is a face model that supports precise local editing and, when paired with a transformer encoder, sets a new state of the art on facial scan registration.

## The mesh problem nobody has quite solved

3D face pipelines for VR, telepresence, and avatars all hit the same tension. 3D Morphable Models built on triangle meshes have remained the default in academia and industry, but they lock you into a fixed topology and a low-dimensional parameter space that struggles with person-specific detail. Newer implicit approaches (signed distance fields decoded by MLPs) capture fine geometry but lack semantic correspondence between faces and require Marching Cubes to extract a mesh. B-spline surfaces from the CAD world offer a middle ground with smooth, editable control lattices, though their limited control points can't represent something as complex as a face.

CUBE tries to keep the useful bits from all three worlds: the semantic correspondence of a template mesh, the expressivity of learned features, and the localized editability of a B-spline.

_Preview only. Full text requires an AlphaSignal Pro account._

## Sources

- [Representing 3D Faces with Learnable B-Spline Volumes](https://arxiv.org/abs/2604.12894)
- [CUBE: Full HTML paper](https://arxiv.org/html/2604.12894)
- [Efficient Learning on Point Clouds with Basis Point Sets](https://arxiv.org/abs/1908.09186)