UniPercept: Towards Unified Perceptual-Level Image Understanding across Aesthetics, Quality, Structure, and Texture
University of Science and Technology of China · Peking University · Shanghai Jiaotong University · Tsinghua University, Tsinghua University · Nanjing university · The University of Sydney · Sun Yat-sen University · Shenzhen Institute of Advanced Technology, Chinese Academy of Sciences · Shanghai Artificial Intelligence Laboratory · Shenzhen Institutes of Advanced Technology, Chinese Academy of Sciences · Shanghai Aritifcal Intelligence Laboratory
PDF 由论文原始站点提供，PaperCompass 不保存论文文件。
摘要
Multimodal large language models (MLLMs) have achieved remarkable progress in visual understanding tasks such as visual grounding, segmentation, and captioning. However, their ability to perceive perceptual-level image features remains limited. In this work, we present UniPercept-Bench, a unified framework for perceptual-level image understanding across three key domains: Aesthetics, Quality, Structure and Texture. We establish a hierarchical definition system and construct large-scale datasets to evaluate perceptual-level image understanding. Based on this foundation, we develop a strong baseline UniPercept trained via Domain-Adaptive Pre-Training and Task-Aligned RL, enabling robust generalization across both Visual Rating (VR) and Visual Question Answering (VQA) tasks. UniPercept outperforms existing MLLMs on perceptual-level image understanding and can serve as a plug-and-play reward model for text-to-image generation. This work defines perceptual-level image understanding in the era of MLLMs and, through the introduction of a comprehensive benchmark together with a strong baseline, provides a solid foundation for advancing perceptual-level multimodal image understanding.