arrow
返回

Enhancing Query Formulation for Universal Image Segmentation

delete2024-03-14
delete2
delete
OA
AI
Y
Yipeng Qu
J
Joohee Kim *
DOI:10.3390/s24061879delete
delete原文链接
delete分享
delete收藏
查看原文
摘要

摘要

En 中文
Recent advancements in image segmentation have been notably driven by Vision Transformers. These transformer-based models offer one versatile network structure capable of handling a variety of segmentation tasks. Despite their effectiveness, the pursuit of enhanced capabilities often leads to more intricate architectures and greater computational demands. OneFormer has responded to these challenges by introducing a query-text contrastive learning strategy active during training only. However, this approach has not completely addressed the inefficiency issues in text generation and the contrastive loss computation. To solve these problems, we introduce Efficient Query Optimizer (EQO), an approach that efficiently utilizes multi-modal data to refine query optimization in image segmentation. Our strategy significantly reduces the complexity of parameters and computations by distilling inter-class and inter-task information from an image into a single template sentence. Furthermore, we propose a novel attention-based contrastive loss. It is designed to facilitate a one-to-many matching mechanism in the loss computation, which helps object queries learn more robust representations. Beyond merely reducing complexity, our model demonstrates superior performance compared to OneFormer across all three segmentation tasks using the Swin-T backbone. Our evaluations on the ADE20K dataset reveal that our model outperforms OneFormer in multiple metrics: by 0.2% in mean Intersection over Union (mIoU), 0.6% in Average Precision (AP), and 0.8% in Panoptic Quality (PQ). These results highlight the efficacy of our model in advancing the field of image segmentation.
Keyword:
computer vision
image segmentation
semantic segmentation
panoptic segmentation
transformer
AI总结

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

Sensors 封面图
Sensors
IF:
3.5
论文数:
7.2W
被引数:
20.9W

机构

I
Illinois Institute of Technology
学者数:
3.8K
论文数: 3.9K
被引数: 4.2K
引用论文

引用论文

NoSQL Systems for Big Data Management
err2014-06-01
err0
PREAI
errVenkat N. Gudivada; Dhana Rao; Vijay V. Raghavan
err分享
err收藏
Systematic review of systematic reviews of non-pharmacological interventions to treat behavioural disturbances in older patients with dementia. The SENATOR-OnTop series
err2017-03-16
err0
errOAAI
errIosief Abraha; Joseph M Rimland; Fabiana Mirella Trotta; Giuseppina Dell'Aquila; Alfonso Cruz-Jentoft; Mirko Petrovic; Adalsteinn Gudmundsson; Roy Soiza; Denis O'Mahony; Antonio Guaita; Antonio Cherubini
err分享
err收藏
err分享
err收藏
Using SEPIC Topology for Improving Power Factor in Distributed Power Supply Systems
err2015-09-22
err0
PREAI
errJ. Sebastián; J. Uceda; J.A. Cobos; J. Arau
err分享
err收藏
Effects of obesity on perception of ability and perception of body image in Portuguese children and adolescents
err2016-01-01
err0
errOAAI
errMiguel Peralta; Adilson Marques; João Martins; Hugo Sarmento; Francisco Carreiro da Costa
err分享
err收藏
Oxidative damage to fibronectin
err1991-02-01
err0
PREAI
errMargret C.M. Vissers; Christine C. Winterbourn
err分享
err收藏
Clustering Image Search Results by Entity Disambiguation
err2014-01-01
err0
errOAAI
errKaiqi Zhao; Zhiyuan Cai; Qingyu Sui; Enxun Wei; Kenny Q. Zhu
err分享
err收藏
学者 查看更多内容