Self-regularized prototypical network for few-shot semantic segmentation

doi:10.1016/j.patcog.2022.109018

2023-01-01

Henghui Ding

Hui Zhang

蒋旭东 (Xudong Jiang) *

DOI：10.1016/j.patcog.2022.109018

原文链接

查看原文

摘要

En 中文

The deep CNNs in image semantic segmentation typically require a large number of densely-annotated images for training and have difficulties in generalizing to unseen object categories. Therefore, few -shot segmentation has been developed to perform segmentation with just a few annotated examples. In this work, we tackle the few-shot segmentation using a self-regularized prototypical network (SRPNet) based on prototype extraction for better utilization of the support information. The proposed SRPNet ex-tracts class-specific prototype representations from support images and generates segmentation masks for query images by a distance metric -the fidelity. A direct yet effective prototype regularization on support set is proposed in SRPNet, in which the generated prototypes are evaluated and regularized on the support set itself. The extent to which the generated prototypes restore the support mask imposes an upper limit on performance. The performance on the query set should never exceed the upper limit no matter how complete the knowledge is generalized from support set to query set. With the specific prototype regularization, SRPNet fully exploits knowledge from the support and offers high-quality proto-types that are representative for each semantic class and meanwhile discriminative for different classes. The query performance is further improved by an iterative query inference (IQI) module that combines a set of regularized prototypes. Our proposed SRPNet achieves new state-of-art performance on 1-shot and 5-shot segmentation benchmarks.(c) 2022 Elsevier Ltd. All rights reserved.

Keyword:

Few-shot segmentation

Prototype

Prototypical network

Self-regularized

Non-parametric distance fidelity

Iterative query inference

SRPNet

CNN

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

Pattern Recognition

IF:

7.6

论文数:

1.3W

被引数:

4.5W

机构

Nanyang Technological University

学者数:

4.9W

论文数: 4.8W

被引数: 8.1W

引用论文

Recent advances in convolutional neural networks 卷积神经网络的最新进展

PATTERN RECOGNITION

IF7.6

2018-05-01

3.8K

OAAI

Gu, Jiuxiang; Wang, Zhenhua; Kuen, Jason; Ma, Lianyang; Shahroudy, Amir; Shuai, Bing; Liu, Ting; Wang, Xingxing; Wang, Gang; Cai, Jianfei; Chen, Tsuhan

Toward Achieving Robust Low-Level and High-Level Scene Parsing

IEEE TRANSACTIONS ON IMAGE PROCESSING

IF13.7

2019-03-01

OAAI

Shuai, Bing; Ding, Henghui; Liu, Ting; Wang, Gang; Jiang, Xudong

A deep one-shot network for query-based logo retrieval

PATTERN RECOGNITION

IF7.6

2019-12-01

OAAI

Bhunia, Ayan Kumar; Bhunia, Ankan Kumar; Ghose, Shuvozit; Das, Abhirup; Roy, Partha Pratim; Pal, Umapada