Learning to Look before Learning to Like: Incorporating Human Visual Cognition into Aesthetic Quality Assessment

2026-04-17

AI Summary

This work addresses a critical limitation in existing automatic aesthetic quality assessment (AQA) methods, which treat images as static pixel vectors and neglect the dynamic nature of human visual exploration that underpins aesthetic judgment. To bridge this gap, the authors propose AestheticNet, the first framework to explicitly incorporate human-like visual cognition by integrating a gaze-aligned visual encoder (GAVE)—pretrained on eye-tracking data to model visual attention—with a fixed semantic encoder such as CLIP via cross-attention. This modular, model-agnostic dual-path architecture enables effective fusion of perceptual attention and semantic understanding. Experiments demonstrate that AestheticNet significantly outperforms purely semantic baselines, and its visual attention module functions as a versatile corrective component that can be flexibly adapted to various AQA backbone networks, thereby validating the necessity and efficacy of human-inspired visual attention in computational aesthetic evaluation.

Technology Category

• Computer Vision: Generative Adversarial Networks (GANs) for Vision
• Cognitive Modeling & Cognitive Systems: Affective Computing
• Humans and AI: Brain-Sensing and Analysis

Application Category

• Economics, Online Markets and Human Computation: LLM based quality controls for crowd work
• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Responsible Web: Human-perceived consequences of algorithmic deployment on the web

Abstract

Automated Aesthetic Quality Assessment (AQA) treats images primarily as static pixel vectors, aligning predictions with human-rating scores largely through semantic perception. However, this paradigm diverges from human aesthetic cognition, which arises from dynamic visual exploration shaped by scanning paths, processing fluency, and the interplay between bottom-up salience and top-down intention. We introduce AestheticNet, a novel cognitive-inspired AQA paradigm that integrates human-like visual cognition and semantic perception with a two-pathway architecture. The visual attention pathway, implemented as a gaze-aligned visual encoder (GAVE) pre-trained offline on eye-tracking data using resource-efficient contrast gaze alignment, models attention from human vision system. This pathway augments the semantic pathway, which uses a fixed semantic encoder such as CLIP, through cross-attention fusion. Visual attention provides a cognitive prior reflecting foreground/background structure, color cascade, brightness, and lighting, all of which are determinants of aesthetic perception beyond semantics. Experiments validated by hypothesis testing show a consistent improvement over the semantic-alone baselines, and demonstrate the gaze module as a model-agnostic corrector compatible with diverse AQA backbones, supporting the necessity and modularity of human-like visual cognition for AQA. Our code is available at https://github.com/keepgallop/AestheticNet.

Problem

• Aesthetic Quality Assessment
• Visual Cognition
• Human Gaze
• Semantic Perception
• Attention Mechanism

Innovation

• visual attention
• gaze alignment
• aesthetic quality assessment
• cognitive-inspired model
• two-pathway architecture