DISN: Deep Implicit Surface Network for High-quality Single-view 3D Reconstruction
This work addresses the challenge of capturing fine details like holes and thin structures in 3D reconstruction from single images, which is important for applications in computer vision and graphics, though it appears incremental as it builds on existing implicit surface methods with a novel feature combination.
The paper tackles the problem of reconstructing 3D shapes from single-view images by introducing DISN, a Deep Implicit Surface Network that predicts signed distance fields to generate high-quality, detail-rich 3D meshes, achieving state-of-the-art performance on various shape categories from synthetic and real images.
Reconstructing 3D shapes from single-view images has been a long-standing research problem. In this paper, we present DISN, a Deep Implicit Surface Network which can generate a high-quality detail-rich 3D mesh from an 2D image by predicting the underlying signed distance fields. In addition to utilizing global image features, DISN predicts the projected location for each 3D point on the 2D image, and extracts local features from the image feature maps. Combining global and local features significantly improves the accuracy of the signed distance field prediction, especially for the detail-rich areas. To the best of our knowledge, DISN is the first method that constantly captures details such as holes and thin structures present in 3D shapes from single-view images. DISN achieves the state-of-the-art single-view reconstruction performance on a variety of shape categories reconstructed from both synthetic and real images. Code is available at https://github.com/xharlie/DISN The supplementary can be found at https://xharlie.github.io/images/neurips_2019_supp.pdf