学习编写用于图像情感分类的多样化提示语

IF 17.3 3区 计算机科学 Q1 COMPUTER SCIENCE, SOFTWARE ENGINEERING
Sinuo Deng, Lifang Wu, Ge Shi, Lehao Xing, Meng Jian, Ye Xiang, Ruihai Dong
{"title":"学习编写用于图像情感分类的多样化提示语","authors":"Sinuo Deng, Lifang Wu, Ge Shi, Lehao Xing, Meng Jian, Ye Xiang, Ruihai Dong","doi":"10.1007/s41095-023-0389-6","DOIUrl":null,"url":null,"abstract":"<p>Image emotion classification (IEC) aims to extract the abstract emotions evoked in images. Recently, language-supervised methods such as contrastive language-image pretraining (CLIP) have demonstrated superior performance in image understanding. However, the underexplored task of IEC presents three major challenges: a tremendous training objective gap between pretraining and IEC, shared suboptimal prompts, and invariant prompts for all instances. In this study, we propose a general framework that effectively exploits the language-supervised CLIP method for the IEC task. First, a prompt-tuning method that mimics the pretraining objective of CLIP is introduced, to exploit the rich image and text semantics associated with CLIP. Subsequently, instance-specific prompts are automatically composed, conditioning them on the categories and image content of instances, diversifying the prompts, and thus avoiding suboptimal problems. Evaluations on six widely used affective datasets show that the proposed method significantly outperforms state-of-the-art methods (up to 9.29% accuracy gain on the EmotionROI dataset) on IEC tasks with only a few trained parameters. The code is publicly available at https://github.com/dsn0w/PT-DPC/for research purposes.\n</p>","PeriodicalId":37301,"journal":{"name":"Computational Visual Media","volume":null,"pages":null},"PeriodicalIF":17.3000,"publicationDate":"2024-04-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":"0","resultStr":"{\"title\":\"Learning to compose diversified prompts for image emotion classification\",\"authors\":\"Sinuo Deng, Lifang Wu, Ge Shi, Lehao Xing, Meng Jian, Ye Xiang, Ruihai Dong\",\"doi\":\"10.1007/s41095-023-0389-6\",\"DOIUrl\":null,\"url\":null,\"abstract\":\"<p>Image emotion classification (IEC) aims to extract the abstract emotions evoked in images. Recently, language-supervised methods such as contrastive language-image pretraining (CLIP) have demonstrated superior performance in image understanding. However, the underexplored task of IEC presents three major challenges: a tremendous training objective gap between pretraining and IEC, shared suboptimal prompts, and invariant prompts for all instances. In this study, we propose a general framework that effectively exploits the language-supervised CLIP method for the IEC task. First, a prompt-tuning method that mimics the pretraining objective of CLIP is introduced, to exploit the rich image and text semantics associated with CLIP. Subsequently, instance-specific prompts are automatically composed, conditioning them on the categories and image content of instances, diversifying the prompts, and thus avoiding suboptimal problems. Evaluations on six widely used affective datasets show that the proposed method significantly outperforms state-of-the-art methods (up to 9.29% accuracy gain on the EmotionROI dataset) on IEC tasks with only a few trained parameters. The code is publicly available at https://github.com/dsn0w/PT-DPC/for research purposes.\\n</p>\",\"PeriodicalId\":37301,\"journal\":{\"name\":\"Computational Visual Media\",\"volume\":null,\"pages\":null},\"PeriodicalIF\":17.3000,\"publicationDate\":\"2024-04-26\",\"publicationTypes\":\"Journal Article\",\"fieldsOfStudy\":null,\"isOpenAccess\":false,\"openAccessPdf\":\"\",\"citationCount\":\"0\",\"resultStr\":null,\"platform\":\"Semanticscholar\",\"paperid\":null,\"PeriodicalName\":\"Computational Visual Media\",\"FirstCategoryId\":\"94\",\"ListUrlMain\":\"https://doi.org/10.1007/s41095-023-0389-6\",\"RegionNum\":3,\"RegionCategory\":\"计算机科学\",\"ArticlePicture\":[],\"TitleCN\":null,\"AbstractTextCN\":null,\"PMCID\":null,\"EPubDate\":\"\",\"PubModel\":\"\",\"JCR\":\"Q1\",\"JCRName\":\"COMPUTER SCIENCE, SOFTWARE ENGINEERING\",\"Score\":null,\"Total\":0}","platform":"Semanticscholar","paperid":null,"PeriodicalName":"Computational Visual Media","FirstCategoryId":"94","ListUrlMain":"https://doi.org/10.1007/s41095-023-0389-6","RegionNum":3,"RegionCategory":"计算机科学","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":null,"EPubDate":"","PubModel":"","JCR":"Q1","JCRName":"COMPUTER SCIENCE, SOFTWARE ENGINEERING","Score":null,"Total":0}
引用次数: 0

摘要

图像情感分类(IEC)旨在提取图像中唤起的抽象情感。最近,对比语言-图像预训练(CLIP)等语言监督方法在图像理解方面表现出了卓越的性能。然而,未被充分探索的 IEC 任务面临着三大挑战:预训练与 IEC 之间巨大的训练目标差距、共享次优提示以及所有实例的不变提示。在本研究中,我们提出了一个通用框架,可有效利用语言监督 CLIP 方法来完成 IEC 任务。首先,我们引入了一种模仿 CLIP 预训练目标的提示调整方法,以利用与 CLIP 相关的丰富图像和文本语义。随后,根据实例的类别和图像内容自动生成针对特定实例的提示,使提示多样化,从而避免出现次优问题。在六个广泛使用的情感数据集上进行的评估表明,在 IEC 任务上,只需少量训练参数,所提出的方法就能明显优于最先进的方法(在 EmotionROI 数据集上的准确率提高了 9.29%)。代码可在 https://github.com/dsn0w/PT-DPC/for 研究网站上公开获取。
本文章由计算机程序翻译,如有差异,请以英文原文为准。

Learning to compose diversified prompts for image emotion classification

Learning to compose diversified prompts for image emotion classification

Image emotion classification (IEC) aims to extract the abstract emotions evoked in images. Recently, language-supervised methods such as contrastive language-image pretraining (CLIP) have demonstrated superior performance in image understanding. However, the underexplored task of IEC presents three major challenges: a tremendous training objective gap between pretraining and IEC, shared suboptimal prompts, and invariant prompts for all instances. In this study, we propose a general framework that effectively exploits the language-supervised CLIP method for the IEC task. First, a prompt-tuning method that mimics the pretraining objective of CLIP is introduced, to exploit the rich image and text semantics associated with CLIP. Subsequently, instance-specific prompts are automatically composed, conditioning them on the categories and image content of instances, diversifying the prompts, and thus avoiding suboptimal problems. Evaluations on six widely used affective datasets show that the proposed method significantly outperforms state-of-the-art methods (up to 9.29% accuracy gain on the EmotionROI dataset) on IEC tasks with only a few trained parameters. The code is publicly available at https://github.com/dsn0w/PT-DPC/for research purposes.

求助全文
通过发布文献求助,成功后即可免费获取论文全文。 去求助
来源期刊
Computational Visual Media
Computational Visual Media Computer Science-Computer Graphics and Computer-Aided Design
CiteScore
16.90
自引率
5.80%
发文量
243
审稿时长
6 weeks
期刊介绍: Computational Visual Media is a peer-reviewed open access journal. It publishes original high-quality research papers and significant review articles on novel ideas, methods, and systems relevant to visual media. Computational Visual Media publishes articles that focus on, but are not limited to, the following areas: • Editing and composition of visual media • Geometric computing for images and video • Geometry modeling and processing • Machine learning for visual media • Physically based animation • Realistic rendering • Recognition and understanding of visual media • Visual computing for robotics • Visualization and visual analytics Other interdisciplinary research into visual media that combines aspects of computer graphics, computer vision, image and video processing, geometric computing, and machine learning is also within the journal''s scope. This is an open access journal, published quarterly by Tsinghua University Press and Springer. The open access fees (article-processing charges) are fully sponsored by Tsinghua University, China. Authors can publish in the journal without any additional charges.
×
引用
GB/T 7714-2015
复制
MLA
复制
APA
复制
导出至
BibTeX EndNote RefMan NoteFirst NoteExpress
×
提示
您的信息不完整,为了账户安全,请先补充。
现在去补充
×
提示
您因"违规操作"
具体请查看互助需知
我知道了
×
提示
确定
请完成安全验证×
copy
已复制链接
快去分享给好友吧!
我知道了
右上角分享
点击右上角分享
0
联系我们:info@booksci.cn Book学术提供免费学术资源搜索服务,方便国内外学者检索中英文文献。致力于提供最便捷和优质的服务体验。 Copyright © 2023 布克学术 All rights reserved.
京ICP备2023020795号-1
ghs 京公网安备 11010802042870号
Book学术文献互助
Book学术文献互助群
群 号:481959085
Book学术官方微信