EmoFeedback$^2$: Reinforcement of Continuous Emotional Image Generation via LVLM-based Reward and Textual Feedback

TL;DR

EmoFeedback$^2$ introduces a reinforcement learning framework utilizing LVLM-based reward and textual feedback to improve continuous emotional image generation, achieving higher emotional fidelity and quality.

Contribution

The paper presents a novel LVLM-based reinforcement learning approach that incorporates emotion-aware rewards and iterative textual feedback for enhanced emotional image generation.

Findings

Outperforms existing methods in emotional fidelity

Generates high-quality images aligned with target emotions

Effectively maintains emotional continuity in images

Abstract

Continuous emotional image generation (C-EICG) is emerging rapidly due to its ability to produce images aligned with both user descriptions and continuous emotional values. However, existing approaches lack emotional feedback from generated images, limiting the control of emotional continuity. Additionally, their simple alignment between emotions and naively generated texts fails to adaptively adjust emotional prompts according to image content, leading to insufficient emotional fidelity. To address these concerns, we propose a novel generation-understanding-feedback reinforcement paradigm (EmoFeedback$^2$) for C-EICG, which exploits the reasoning capability of the fine-tuned large vision-language model (LVLM) to provide reward and textual feedback for generating high-quality images with continuous emotions. Specifically, we introduce an emotion-aware reward feedback strategy, where the LVLM evaluates the emotional values of generated images and computes the reward against target emotions, guiding the reinforcement fine-tuning of the generative model and enhancing the emotional continuity of images. Furthermore, we design a self-promotion textual feedback framework, in which the LVLM iteratively analyzes the emotional content of generated images and adaptively produces refinement suggestions for the next-round prompt, improving the emotional fidelity with fine-grained content. Extensive experimental results demonstrate that our approach effectively generates high-quality images with the desired emotions, outperforming existing state-of-the-art methods in our custom dataset. The code and dataset will be released soon.