arrow
Return

Rethinking Rotation-Invariant Recognition of Fine-Grained Shapes From the Perspective of Contour Points

delete2025-11-21
delete0
PRE
AI
Y
Yanjie Xu
Y
Yunzhi Chen
H
Handing Xu
T
Tianmu Wang
Y
Yaguan Li
聂振国 cover
聂振国 (Zhenguo Nie)
DOI:10.1109/TPAMI.2025.3635480delete
deleteOriginal
deleteShare
deleteSave
Abstract

Abstract

En 中文
Rotation-invariant recognition of shapes is a common challenge in computer vision. Recent approaches have significantly improved the accuracy of rotation-invariant recognition by encoding the rotational invariance of shapes as hand-crafted image features and introducing deep neural networks. However, the methods based on pixels have too much redundant information, and the critical geometric information is prone to early leakage, resulting in weak rotation-invariant recognition of fine-grained shapes. In this paper, we reconsider the shape recognition problem from the perspective of contour points rather than pixels. We propose an anti-noise rotation-invariant convolution module based on contour geometric aware for fine-grained shape recognition. The module divides the shape contour into multiple local geometric regions (LGA), where we implement finer-grained rotation-invariant coding in terms of point topological relations. We provide a deep network composed of five such cascaded modules for classification and retrieval experiments. The results show that our method exhibits excellent performance in rotation-invariant recognition of fine-grained shapes. In addition, we demonstrate that our method is robust to contour noise and the rotation centers.
Keywords:
Fine-grained shape recognition
rotation invariance
deep learning
contour geometric aware

Journal

IEEE Transactions on Pattern Analysis and Machine Intelligence cover
IEEE Transactions on Pattern Analysis and Machine Intelligence
IF:
18.6
Papers:
2.1K
Citations:
9.8W

Organization

T
Tsinghua University
Scholars:
8.6K
Papers: 4.1K
Citations: 17.7W
T
taiyuan university of technology
Scholars:
6.8K
Papers: 2.1K
Citations: 0
Cited Papers

Cited Papers

errShare
errSave
BodyParts3D: 3D structure database for anatomical concepts
err2009-01-01
err159
errOAAI
errMitsuhashi, Nobutaka; Fujieda, Kaori; Tamura, Takuro; Kawamoto, Shoko; Takagi, Toshihisa; Okubo, Kousaku
errShare
errSave
Integration of operator-validated contours in deformable image registration for dose accumulation in radiotherapy
err2023-07-01
err6
errOAAI
errBosma, Lando S.; Ries, Mario; de Senneville, Baudouin Denis; Raaymakers, Bas W.; Zachiu, Cornel
errShare
errSave
Graph-based geometric structure line parsing
err2024-05-01
err0
PREAI
errLi, Feng; Li, Gang; He, Bin; Lu, Ping; Cheng, Bin
errShare
errSave
Estimating 3D Hand Poses and Shapes from Silhouettes
err
err0
PREAI
errChang,Li-Jen; Liao,Yu-Cheng; Lin,Chia-Hui; Yang-Mao,Shys-Fang; Chen,Hwann-Tzong
errShare
errSave
Real-Time Deep Pose Estimation With Geodesic Loss for Image-to-Template Rigid Registration
err2019-02-01
err0
errOAAI
errSeyed Sadegh Mohseni Salehi; Shadab Khan; Deniz Erdogmus; Ali Gholipour
errShare
errSave
Transformation-invariant Gabor convolutional networks
err2020-04-17
err0
PREAI
errLei Zhuang; Feipeng Da; Shaoyan Gai; Mengxiang Li
errShare
errSave
researcher View more