Pragmatist: Multiview Conditional Diffusion Models for High-Fidelity 3D Reconstruction from Unposed Sparse Views

TL;DR

Pragmatist introduces a multiview conditional diffusion approach to improve high-fidelity 3D reconstruction from sparse, unposed views by generating complete observations and leveraging geometric priors.

Contribution

The paper proposes a novel pipeline combining multiview diffusion models with a reconstruction network to enhance 3D reconstruction from unposed sparse views, addressing limitations of prior methods.

Findings

Achieves promising results on multiple benchmarks.

Effectively leverages unposed inputs and generative priors.

Improves geometric and textural detail reconstruction.

Abstract

Inferring 3D structures from sparse, unposed observations is challenging due to its unconstrained nature. Recent methods propose to predict implicit representations directly from unposed inputs in a data-driven manner, achieving promising results. However, these methods do not utilize geometric priors and cannot hallucinate the appearance of unseen regions, thus making it challenging to reconstruct fine geometric and textural details. To tackle this challenge, our key idea is to reformulate this ill-posed problem as conditional novel view synthesis, aiming to generate complete observations from limited input views to facilitate reconstruction. With complete observations, the poses of the input views can be easily recovered and further used to optimize the reconstructed object. To this end, we propose a novel pipeline Pragmatist. First, we generate a complete observation of the object via a multiview conditional diffusion model. Then, we use a feed-forward large reconstruction model to obtain the reconstructed mesh. To further improve the reconstruction quality, we recover the poses of input views by inverting the obtained 3D representations and further optimize the texture using detailed input views. Unlike previous approaches, our pipeline improves reconstruction by efficiently leveraging unposed inputs and generative priors, circumventing the direct resolution of highly ill-posed problems. Extensive experiments show that our approach achieves promising performance in several benchmarks.