HybridSDF: Combining Deep Implicit Shapes and Geometric Primitives for 3D Shape Representation and Manipulation

被引：1

作者：

Vasu, Subeesh ^{[1
]}

Talabot, Nicolas ^{[1
]}

Lukoianov, Artem ^{[1
,2
]}

Baque, Pierre ^{[2
]}

Donier, Jonathan ^{[2
]}

Fua, Pascal ^{[1
]}

机构：

[1] Ecole Polytech Fed Lausanne, CVLab, Lausanne, Switzerland

[2] Neural Concept, Lausanne, Switzerland

来源：

2022 INTERNATIONAL CONFERENCE ON 3D VISION, 3DV | 2022年

关键词：

D O I：

10.1109/3DV57658.2022.00072

中图分类号：

TP18 [人工智能理论];

学科分类号：

081104 ; 0812 ; 0835 ; 1405 ;

摘要：

Deep implicit surfaces excel at modeling generic shapes but do not always capture the regularities present in manufactured objects, which is something simple geometric primitives are particularly good at. In this paper, we propose a representation combining latent and explicit parameters that can be decoded into a set of deep implicit and geometric shapes that are consistent with each other. As a result, we can effectively model both complex and highly regular shapes that coexist in manufactured objects. This enables our approach to manipulate 3D shapes in an efficient and precise manner.

引用

页码：617 / 626

页数：10

共 50 条

[31] On the multiscale representation of 2D and 3D shapes
Borgefors, G
Ramella, G
di Baja, GS
Svensson, S
GRAPHICAL MODELS AND IMAGE PROCESSING, 1999, 61 (01): : 44 - 62
[32] Deep Implicit Statistical Shape Models for 3D Lumbar Vertebrae Image Delineation
Ocepek, Domen
Podobnik, Gasper
Ibragimov, Bulat
Vrtovec, Tomaz
MEDICAL IMAGING 2024: IMAGE PROCESSING, 2024, 12926
[33] Semantic Segmentation of Geometric Primitives in Dense 3D Point Clouds
Stanescu, Ana
Fleck, Philipp
Schmalstieg, Dieter
Arth, Clemens
ADJUNCT PROCEEDINGS OF THE 2018 IEEE INTERNATIONAL SYMPOSIUM ON MIXED AND AUGMENTED REALITY (ISMAR), 2018, : 206 - 211
[34] Determining Geometric Primitives for a 3D GIS Easy as 1D, 2D, 3D?
Lonneville, Britt
Stal, Cornelis
De Roo, Berdien
De Wulf, Alain
De Maeyer, Philippe
2015 1ST INTERNATIONAL CONFERENCE ON GEOGRAPHICAL INFORMATION SYSTEMS THEORY, APPLICATIONS AND MANAGEMENT (GISTAM), 2015, : 135 - 140
[35] 3D registration by using an alternative 3D shape representation
Torre-Ferrero, C.
Robla, S.
Sarabia, E. G.
Llata, J. R.
PROCEEDINGS OF THE 7TH WSEAS INTERNATIONAL CONFERENCE ON SIGNAL PROCESSING, COMPUTATIONAL GEOMETRY AND ARTIFICIAL VISION (ISCGAV'-07), 2007, : 97 - +
[36] Geometry curves: A compact representation for 3D shapes
Li, Guo
Liu, Ligang
GRAPHICAL MODELS, 2013, 75 : 265 - 278
[37] Research on 3D Reconstruction Methods for Incomplete Building Point Clouds Using Deep Learning and Geometric Primitives
Ding, Ziqi
Lu, Yuefeng
Shao, Shiwei
Qin, Yong
Lu, Miao
Song, Zhenqi
Sun, Dengkuo
REMOTE SENSING, 2025, 17 (03)
[38] Unsupervised recursive deep fitting of 3D primitives to points
Saporta, Tsahi
Sharf, Andrei
COMPUTERS & GRAPHICS-UK, 2022, 102 : 289 - 299
[39] 3D model segmentation and representation with implicit polynomials
Zheng, Bo
Takamatsu, Jun
Ikeuchi, Katsushi
IEICE TRANSACTIONS ON INFORMATION AND SYSTEMS, 2008, E91D (04) : 1149 - 1158
[40] CurvaNet: Geometric Deep Learning based on Directional Curvature for 3D Shape Analysis
He, Wenchong
Jiang, Zhe
Zhang, Chengming
Sainju, Arpan Man
KDD '20: PROCEEDINGS OF THE 26TH ACM SIGKDD INTERNATIONAL CONFERENCE ON KNOWLEDGE DISCOVERY & DATA MINING, 2020, : 2214 - 2224

← 1 2 3 4 5 →