Learning to Reconstruct 3D Manhattan Wireframes from a Single Image

All Research

Abstract

In this paper, we propose a method to obtain a compact and accurate 3D wireframe representation from a single image by effectively exploiting global structural regularities. Our method trains a convolutional neural network to simultaneously detect salient junctions and straight lines, as well as predict their 3D depth and vanishing points. Compared with the state-of-the-art learning-based wireframe detection methods, our network is much simpler and more unified, leading to better 2D wireframe detection. With global structural priors such as Manhattan assumption, our method further reconstructs a full 3D wireframe model, a compact vector representation suitable for a variety of high-level vision tasks such as AR and CAD. We conduct extensive evaluations on a large synthetic dataset of urban scenes as well as real images. Our code and datasets will be released.

Bibtex

@INPROCEEDINGS{9010693,
author={Y. {Zhou} and H. {Qi} and Y. {Zhai} and Q. {Sun} and Z. {Chen} and L. {Wei} and Y. {Ma}},
booktitle={2019 IEEE/CVF International Conference on Computer Vision (ICCV)},
title={Learning to Reconstruct 3D Manhattan Wireframes From a Single Image},
year={2019},
volume={},
number={},
pages={7697-7706},
doi={10.1109/ICCV.2019.00779}}