Query-guided Prototype Evolution Network for Few-Shot Segmentation

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Cong, Runmin, Xiong, Hang, Chen, Jinpeng, Zhang, Wei, Huang, Qingming, Zhao, Yao
Natura: Preprint
Pubblicazione: 2024
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866910731933319168
author Cong, Runmin
Xiong, Hang
Chen, Jinpeng
Zhang, Wei
Huang, Qingming
Zhao, Yao
author_facet Cong, Runmin
Xiong, Hang
Chen, Jinpeng
Zhang, Wei
Huang, Qingming
Zhao, Yao
contents Previous Few-Shot Segmentation (FSS) approaches exclusively utilize support features for prototype generation, neglecting the specific requirements of the query. To address this, we present the Query-guided Prototype Evolution Network (QPENet), a new method that integrates query features into the generation process of foreground and background prototypes, thereby yielding customized prototypes attuned to specific queries. The evolution of the foreground prototype is accomplished through a \textit{support-query-support} iterative process involving two new modules: Pseudo-prototype Generation (PPG) and Dual Prototype Evolution (DPE). The PPG module employs support features to create an initial prototype for the preliminary segmentation of the query image, resulting in a pseudo-prototype reflecting the unique needs of the current query. Subsequently, the DPE module performs reverse segmentation on support images using this pseudo-prototype, leading to the generation of evolved prototypes, which can be considered as custom solutions. As for the background prototype, the evolution begins with a global background prototype that represents the generalized features of all training images. We also design a Global Background Cleansing (GBC) module to eliminate potential adverse components mirroring the characteristics of the current foreground class. Experimental results on the PASCAL-$5^i$ and COCO-$20^i$ datasets attest to the substantial enhancements achieved by QPENet over prevailing state-of-the-art techniques, underscoring the validity of our ideas.
format Preprint
id arxiv_https___arxiv_org_abs_2403_06488
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle Query-guided Prototype Evolution Network for Few-Shot Segmentation
Cong, Runmin
Xiong, Hang
Chen, Jinpeng
Zhang, Wei
Huang, Qingming
Zhao, Yao
Computer Vision and Pattern Recognition
Previous Few-Shot Segmentation (FSS) approaches exclusively utilize support features for prototype generation, neglecting the specific requirements of the query. To address this, we present the Query-guided Prototype Evolution Network (QPENet), a new method that integrates query features into the generation process of foreground and background prototypes, thereby yielding customized prototypes attuned to specific queries. The evolution of the foreground prototype is accomplished through a \textit{support-query-support} iterative process involving two new modules: Pseudo-prototype Generation (PPG) and Dual Prototype Evolution (DPE). The PPG module employs support features to create an initial prototype for the preliminary segmentation of the query image, resulting in a pseudo-prototype reflecting the unique needs of the current query. Subsequently, the DPE module performs reverse segmentation on support images using this pseudo-prototype, leading to the generation of evolved prototypes, which can be considered as custom solutions. As for the background prototype, the evolution begins with a global background prototype that represents the generalized features of all training images. We also design a Global Background Cleansing (GBC) module to eliminate potential adverse components mirroring the characteristics of the current foreground class. Experimental results on the PASCAL-$5^i$ and COCO-$20^i$ datasets attest to the substantial enhancements achieved by QPENet over prevailing state-of-the-art techniques, underscoring the validity of our ideas.
title Query-guided Prototype Evolution Network for Few-Shot Segmentation
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2403.06488