StdGEN++: A Comprehensive System for Semantic-Decomposed 3D Character Generation

TL;DR

StdGEN++ is a comprehensive system that generates high-fidelity, semantically decomposed 3D characters, enabling flexible editing and downstream applications through innovative reconstruction, surface extraction, and texture decomposition techniques.

Contribution

It introduces a dual-branch semantic-aware reconstruction model and a novel semantic surface extraction formalism for high-quality, editable 3D character generation.

Findings

Achieves state-of-the-art geometric accuracy and semantic disentanglement.

Enables non-destructive editing and physics-based animation.

Reduces memory footprint for high-resolution mesh generation.

Abstract

We present StdGEN++, a novel and comprehensive system for generating high-fidelity, semantically decomposed 3D characters from diverse inputs. Existing 3D generative methods often produce monolithic meshes that lack the structural flexibility required by industrial pipelines in gaming and animation. Addressing this gap, StdGEN++ is built upon a Dual-branch Semantic-aware Large Reconstruction Model (Dual-Branch S-LRM), which jointly reconstructs geometry, color, and per-component semantics in a feed-forward manner. To achieve production-level fidelity, we introduce a novel semantic surface extraction formalism compatible with hybrid implicit fields. This mechanism is accelerated by a coarse-to-fine proposal scheme, which significantly reduces memory footprint and enables high-resolution mesh generation. Furthermore, we propose a video-diffusion-based texture decomposition module that disentangles appearance into editable layers (e.g., separated iris and skin), resolving semantic confusion in facial regions. Experiments demonstrate that StdGEN++ achieves state-of-the-art performance, significantly outperforming existing methods in geometric accuracy and semantic disentanglement. Crucially, the resulting structural independence unlocks advanced downstream capabilities, including non-destructive editing, physics-compliant animation, and gaze tracking, making it a robust solution for automated character asset production.