UniPercept: Towards Unified Perceptual-Level Image Understanding across Aesthetics, Quality, Structure, and Texture
This work addresses the problem of perceptual-level image understanding for researchers and practitioners in computer vision and multimodal AI, though it appears incremental as it builds on existing MLLM frameworks with new datasets and training methods.
The paper tackled the limited ability of multimodal large language models (MLLMs) to perceive perceptual-level image features by introducing UniPercept-Bench, a unified framework for image understanding across aesthetics, quality, structure, and texture, which outperforms existing MLLMs and serves as a plug-and-play reward model for text-to-image generation.
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.