3D Representation Methods: A Survey
Zhengren Wang
TL;DR
The survey maps the landscape of 3D representations by organizing core methods into explicit (voxel grids, meshes, point clouds) and implicit (SDFs, NeRF) categories, while highlighting hybrid approaches such as DMTet and Tri-plane that fuse strengths from multiple paradigms. It reviews foundational works, variant models, and differentiable rendering techniques, and catalogs influential datasets (e.g., ShapeNet, ScanNet, Objaverse) that drive progress across reconstruction, rendering, and scene understanding. The paper identifies key research directions including efficiency, scalability, and cross-domain applications (AR/VR, medical imaging), and discusses how data generation and hybrid representations can expand capabilities for real-time, high-fidelity 3D content. Overall, the work provides a comprehensive framework for comparing 3D representations, informing future research, dataset creation, and practical deployment in graphics, vision, and robotics.
Abstract
The field of 3D representation has experienced significant advancements, driven by the increasing demand for high-fidelity 3D models in various applications such as computer graphics, virtual reality, and autonomous systems. This review examines the development and current state of 3D representation methods, highlighting their research trajectories, innovations, strength and weakness. Key techniques such as Voxel Grid, Point Cloud, Mesh, Signed Distance Function (SDF), Neural Radiance Field (NeRF), 3D Gaussian Splatting, Tri-Plane, and Deep Marching Tetrahedra (DMTet) are reviewed. The review also introduces essential datasets that have been pivotal in advancing the field, highlighting their characteristics and impact on research progress. Finally, we explore potential research directions that hold promise for further expanding the capabilities and applications of 3D representation methods.