Pix2Surf: Learning Parametric 3D Surface Models of Objects from Images

TL;DR

This paper introduces a neural network approach for generating consistent, high-quality 3D parametric surface models of objects from images, improving multi-view shape reconstruction and texture transfer.

Contribution

It presents a novel neural network architecture that produces continuous, view-consistent 3D surfaces with accurate pixel-to-surface correspondences, surpassing prior discrete and inconsistent methods.

Findings

Outperforms previous shape reconstruction methods quantitatively.

Produces high-quality, multi-view consistent 3D surfaces.

Enables detailed texture transfer with accurate geometry.

Abstract

We investigate the problem of learning to generate 3D parametric surface representations for novel object instances, as seen from one or more views. Previous work on learning shape reconstruction from multiple views uses discrete representations such as point clouds or voxels, while continuous surface generation approaches lack multi-view consistency. We address these issues by designing neural networks capable of generating high-quality parametric 3D surfaces which are also consistent between views. Furthermore, the generated 3D surfaces preserve accurate image pixel to 3D surface point correspondences, allowing us to lift texture information to reconstruct shapes with rich geometry and appearance. Our method is supervised and trained on a public dataset of shapes from common object categories. Quantitative results indicate that our method significantly outperforms previous work, while qualitative results demonstrate the high quality of our reconstructions.