Affective Image Manipulation (AIM) aims to alter an image's emotional impact by adjusting multiple visual elements to evoke specificthis http URLAIM is inherently complex, necessitating a collaborative approach that involves identifying semantic cues within source images, manipulating these elements to elicit desired emotional responses, and verifying that the combined adjustments successfully evoke the targetthis http URLaddress these challenges, we introduce EmoAgent, the first multi-agent collaboration framework for AIM. By emulating the cognitive behaviors of a human painter, EmoAgent incorporates three specialized agents responsible for planning, editing, and critical evaluation. Furthermore, we develop an emotion-factor knowledge retriever, a decision-making tree space, and a tool library to enhance EmoAgent's effectiveness in handling AIM. Experiments demonstrate that the proposed multi-agent framework outperforms existing methods, offering more reasonable and effective emotional expression.
View on arXiv@article{mao2025_2503.11290,
title={ EmoAgent: Multi-Agent Collaboration of Plan, Edit, and Critic, for Affective Image Manipulation },
author={ Qi Mao and Haobo Hu and Yujie He and Difei Gao and Haokun Chen and Libiao Jin },
journal={arXiv preprint arXiv:2503.11290},
year={ 2025 }
}