Occlusion-Aware Instance Segmentation Via BiLayer Network Architectures
IEEE Transactions on Pattern Analysis and Machine Intelligence · 2023-08 · 45 (8) · 10197-10211
DOI: 10.1109/tpami.2023.3246174 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Pattern Analysis and Machine Intelligence, 2023-08. Verfasst von Lei Ke und 2 weiteren Autor:innen. Hauptthemen laut MEDLINE: Algorithms, Image Processing, Computer-Assisted. Die Arbeit wurde bisher 39-mal zitiert (Crossref) und nennt 99 Literaturangaben. Gefördert von: Research Grant Council of the Hong Kong SAR.
Abstract (Originalsprache)
Segmenting highly-overlapping image objects is challenging, because there is typically no distinction between real object contours and occlusion boundaries on images. Unlike previous instance segmentation methods, we model image formation as a composition of two overlapping layers, and propose Bilayer Convolutional Network (BCNet), where the top layer detects occluding objects (occluders) and the bottom layer infers partially occluded instances (occludees). The explicit modeling of occlusion relationship with bilayer structure naturally decouples the boundaries of both the occluding and occluded instances, and considers the interaction between them during mask regression. We investigate the efficacy of bilayer structure using two popular convolutional network designs, namely, Fully Convolutional Network (FCN) and Graph Convolutional Network (GCN). Further, we formulate bilayer decoupling using the vision transformer (ViT), by representing instances in the image as separate learnable occluder and occludee queries. Large and consistent improvements using one/two-stage and query-based object detectors with various backbones and network layer choices validate the generalization ability of bilayer decoupling, as shown by extensive experiments on image instance segmentation benchmarks (COCO, KINS, COCOA) and video instance segmentation benchmarks (YTVIS, OVIS, BDD100 K MOTS), especially for heavy occlusion cases.
Schlagwörter
MeSH-Schlagwörter (MEDLINE)
Förderung
- Research Grant Council of the Hong Kong SAR
Zitieren
APA
Ke, L., Tai, Y. W., & Tang, C. K. (2023). Occlusion-Aware Instance Segmentation Via BiLayer Network Architectures. IEEE Transactions on Pattern Analysis and Machine Intelligence, 45(8), 10197-10211. https://doi.org/10.1109/tpami.2023.3246174
Vancouver
Ke L, Tai YW, Tang CK. Occlusion-Aware Instance Segmentation Via BiLayer Network Architectures. IEEE Transactions on Pattern Analysis and Machine Intelligence. 2023;45(8):10197-10211. doi:10.1109/tpami.2023.3246174
BibTeX
@article{Ke2023,
author = {Ke, Lei and Tai, Yu-Wing and Tang, Chi-Keung},
title = {Occlusion-Aware Instance Segmentation Via BiLayer Network Architectures},
journal = {IEEE Transactions on Pattern Analysis and Machine Intelligence},
year = {2023},
volume = {45},
number = {8},
pages = {10197--10211},
doi = {10.1109/tpami.2023.3246174}
}
EUVIMED