SPAR3D: Stable Point-Aware Reconstruction of 3D Objects from Single Images
This addresses the problem of efficient and accurate 3D reconstruction from single images for applications like computer vision and graphics, representing an incremental improvement by hybridizing existing approaches.
The paper tackles single-image 3D object reconstruction by proposing SPAR3D, a two-stage method that combines regression and generative modeling to efficiently handle occluded regions and maintain output fidelity, achieving state-of-the-art performance with an inference speed of 0.7 seconds.
We study the problem of single-image 3D object reconstruction. Recent works have diverged into two directions: regression-based modeling and generative modeling. Regression methods efficiently infer visible surfaces, but struggle with occluded regions. Generative methods handle uncertain regions better by modeling distributions, but are computationally expensive and the generation is often misaligned with visible surfaces. In this paper, we present SPAR3D, a novel two-stage approach aiming to take the best of both directions. The first stage of SPAR3D generates sparse 3D point clouds using a lightweight point diffusion model, which has a fast sampling speed. The second stage uses both the sampled point cloud and the input image to create highly detailed meshes. Our two-stage design enables probabilistic modeling of the ill-posed single-image 3D task while maintaining high computational efficiency and great output fidelity. Using point clouds as an intermediate representation further allows for interactive user edits. Evaluated on diverse datasets, SPAR3D demonstrates superior performance over previous state-of-the-art methods, at an inference speed of 0.7 seconds. Project page with code and model: https://spar3d.github.io