NeuMesh++: Towards Versatile and Efficient Volumetric Editing with Disentangled Neural Mesh-based Implicit Field

2026-06-17

AI Summary

This work addresses the limited editability of existing neural rendering methods, which struggle to flexibly manipulate geometry, texture, and semantics. The authors propose a mesh-based neural implicit field representation that disentangles these components through vertex-level decoupled encoding. Their approach incorporates local spatial parameterization, learnable vertex albedo, and a semantic-guided region selection mechanism. This formulation enables mesh-guided geometric editing, localized texture replacement or painting, and semantic-driven manipulation. Evaluated on both real-world and synthetic datasets, the method achieves high-fidelity and diverse 3D content editing, significantly improving representational quality and editing flexibility.

Abstract

Recently neural implicit rendering techniques have evolved rapidly and demonstrated significant advantages in novel view synthesis and 3D scene reconstruction. However, existing neural rendering methods for editing purposes offer limited functionalities, e.g., rigid transformation and category-specific editing. In this paper, we present a novel mesh-based representation by encoding the neural radiance field with disentangled geometry, texture, and semantic codes on mesh vertices, which empowers a set of efficient and comprehensive editing functionalities, including mesh-guided geometry editing, designated texture editing with texture swapping, filling and painting operations, and semantic-guided editing. To this end, we develop several techniques including a novel local space parameterization to enhance rendering quality and training stability, a learnable modification color on vertex to improve the fidelity of texture editing, a spatial-aware optimization strategy to realize precise texture editing, and a semantic-aided region selection to ease the laborious annotation of implicit field editing. Extensive experiments and editing examples on both real and synthetic datasets demonstrate the superiority of our method on representation quality and editing ability. Project page: https://zju3dv.github.io/neumeshplusplus/

Problem

• neural implicit rendering
• volumetric editing
• 3D scene editing
• mesh-based representation
• disentangled representation

Innovation

• disentangled representation
• neural implicit field
• mesh-based editing
• texture editing
• semantic-guided editing