NoPose-NeuS: Jointly Optimizing Camera Poses with Neural Implicit Surfaces for Multi-view Reconstruction

TL;DR

NoPose-NeuS introduces a method that jointly optimizes camera poses and neural implicit surfaces for multi-view reconstruction, improving pose accuracy and surface quality without prior pose information.

Contribution

It extends NeuS by integrating camera pose optimization into the neural surface reconstruction process using pose encoding and novel loss functions.

Findings

Achieves high-quality surface reconstruction with 0.89 mean Chamfer distance.

Estimates camera poses accurately without prior pose data.

Maintains robustness across complex geometries and non-Lambertian surfaces.

Abstract

Learning neural implicit surfaces from volume rendering has become popular for multi-view reconstruction. Neural surface reconstruction approaches can recover complex 3D geometry that are difficult for classical Multi-view Stereo (MVS) approaches, such as non-Lambertian surfaces and thin structures. However, one key assumption for these methods is knowing accurate camera parameters for the input multi-view images, which are not always available. In this paper, we present NoPose-NeuS, a neural implicit surface reconstruction method that extends NeuS to jointly optimize camera poses with the geometry and color networks. We encode the camera poses as a multi-layer perceptron (MLP) and introduce two additional losses, which are multi-view feature consistency and rendered depth losses, to constrain the learned geometry for better estimated camera poses and scene surfaces. Extensive experiments on the DTU dataset show that the proposed method can estimate relatively accurate camera poses, while maintaining a high surface reconstruction quality with 0.89 mean Chamfer distance.