DualPrim: Compact 3D Reconstruction with Positive and Negative Primitives

TL;DR

DualPrim introduces a novel 3D reconstruction method using positive and negative primitives, achieving high accuracy with compact, structured, and editable models suitable for downstream applications.

Contribution

It presents a new additive-subtractive primitive framework that enhances shape representation and integrates with differentiable rendering for end-to-end learning.

Findings

State-of-the-art reconstruction accuracy

Produces compact and structured 3D models

Better suited for downstream editing and reuse

Abstract

Neural reconstructions often trade structure for fidelity, yielding dense and unstructured meshes with irregular topology and weak part boundaries that hinder editing, animation, and downstream asset reuse. We present DualPrim, a compact and structured 3D reconstruction framework. Unlike additive-only implicit or primitive methods, DualPrim represents shapes with positive and negative superquadrics: the former builds the bases while the latter carves local volumes through a differentiable operator, enabling topology-aware modeling of holes and concavities. This additive-subtractive design increases the representational power without sacrificing compactness or differentiability. We embed DualPrim in a volumetric differentiable renderer, enabling end-to-end learning from multi-view images and seamless mesh export via closed-form boolean difference. Empirically, DualPrim delivers state-of-the-art accuracy and produces compact, structured, and interpretable outputs that better satisfy downstream needs than additive-only alternatives.