arrow
返回

Neural Vector Fields: Generalizing Distance Vector Fields by Codebooks and Zero-Curl Regularization

delete2025-01-01
delete0
delete
OA
AI
X
Xianghui Yang
G
Guosheng Lin
Z
Zhenghao Chen
L
Luping Zhou
DOI:10.1109/tpami.2025.3552684delete
delete原文链接
delete分享
delete收藏
查看原文
摘要

摘要

En 中文
Recent neural networks based surface reconstruction can be roughly divided into two categories, one warping templates explicitly and the other representing 3D surfaces implicitly. To enjoy the advantages of both, we propose a novel 3D representation, Neural Vector Fields (NVF), which adopts the explicit learning process to manipulate meshes and implicit unsigned distance function (UDF) representation to break the barriers in resolution and topology. This is achieved by directly predicting the displacements from surface queries and modeling shapes as Vector Fields, rather than relying on network differentiation to obtain direction fields as most existing UDF-based methods do. In this way, our approach is capable of encoding both the distance and the direction fields so that the calculation of direction fields is differentiation-free, circumventing the non-trivial surface extraction step. Furthermore, building upon NVFs, we propose to incorporate two types of shape codebooks, i.e., NVFs (Lite or Ultra), to promote cross-category reconstruction through encoding cross-object priors. Moreover, we propose a new regularization based on analyzing the zero-curl property of NVFs, and implement this through the fully differentiable framework of our NVF (ultra). We evaluate both NVFs on four surface reconstruction scenarios, including watertight vs non-watertight shapes, category-agnostic reconstruction vs category-unseen reconstruction, category-specific, and cross-domain reconstruction.
AI总结

AI总结

对已上传原文的论文进行重点信息的提取,主要内容包括:简要概述、研究摘要、背景介绍、关键亮点、图文解析、展望与总结。

期刊

IEEE Transactions on Pattern Analysis and Machine Intelligence 封面图
IEEE Transactions on Pattern Analysis and Machine Intelligence
IF:
18.6
论文数:
2.9K
被引数:
10.9W

机构

南
南洋理工大学
学者数:
4.2K
论文数: 2.0K
被引数: 5.3W
T
the university of sydney
学者数:
1.6K
论文数: 678
被引数: 862
引用论文

引用论文

Multi-Garment Net: Learning to Dress 3D People From Images
err2019-10-01
err0
errOAAI
errBharat Bhatnagar; Garvita Tiwari; Christian Theobalt; Gerard Pons-Moll
err分享
err收藏
err分享
err收藏
PIFu: Pixel-Aligned Implicit Function for High-Resolution Clothed Human Digitization
err2019-10-01
err0
errOAAI
errShunsuke Saito; Zeng Huang; Ryota Natsume; Shigeo Morishima; Hao Li; Angjoo Kanazawa
err分享
err收藏
Local Deep Implicit Functions for 3D Shape
err2020-06-01
err0
errOAAI
errKyle Genova; Forrester Cole; Avneesh Sud; Aaron Sarna; Thomas Funkhouser
err分享
err收藏
NASA Neural Articulated Shape Approximation
err2020-11-09
err0
PREAI
errBoyang Deng; J. P. Lewis; Timothy Jeruzalski; Gerard Pons-Moll; Geoffrey Hinton; Mohammad Norouzi; Andrea Tagliasacchi
err分享
err收藏
ShapeFormer: Transformer-based Shape Completion via Sparse Representation
err2022-06-01
err0
errOAAI
errXingguang Yan; Liqiang Lin; Niloy J. Mitra; Dani Lischinski; Daniel Cohen-Or; Hui Huang
err分享
err收藏
学者 查看更多内容