ECCV 2020

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

1 Zhejiang University
2 Stanford University
3 Adobe Research
4 University College London





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.


author = {Lei, Jiahui and Sridhar, Srinath and Guerrero, Paul and Sung, Minhyuk and Mitra, Niloy and Guibas, Leonidas J.},
title = {Pix2Surf: Learning Parametric 3D Surface Models of Objects from Images},
booktitle = {Proceedings of European Conference on Computer Vision ({ECCV})},
url = {},
month = August,
year = {2020}}


We thank the anonymous reviewers for their comments and suggestions. This work was supported by a Vannevar Bush Faculty Fellowship, NSF grant IIS-1763268, grants from the Stanford GRO Program, the SAIL-Toyota Center for AI Research, AWS Machine Learning Awards Program, UCL AI Center, and a gift from the Adobe. We thank Supriya Krishnan for help with the voice over.


Jiahui Lei (

Srinath Sridhar (