Venus: Benchmarking and Empowering Multimodal Large Language Models for Aesthetic Guidance and Cropping

2026-02-27

AI Summary

Existing vision-language models struggle to identify aesthetic issues in images and provide actionable photography and cropping suggestions, limiting their effectiveness in composition optimization tasks. To address this, this work proposes Venus, a framework trained in two stages: first, progressive complexity-based aesthetic question answering enhances the model’s ability to deliver aesthetic guidance; second, chain-of-thought (CoT) reasoning is leveraged to activate its aesthetic cropping capability. We introduce AesGuide, the first large-scale dataset for aesthetic instruction, and develop the first interpretable, interactive multimodal large language model system for aesthetic guidance and cropping. Experiments demonstrate that Venus achieves state-of-the-art performance on both aesthetic guidance and cropping tasks, enabling end-to-end, full-pipeline aesthetic optimization.

Technology Category

• Computer Vision: Large Vision Models
• Planning, Routing, and Scheduling: Planning with Language Models
• Search and Optimization: Learning to Search

Application Category

• Search and Retrieval-Augmented AI: Agentic search
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs

Abstract

The widespread use of smartphones has made photography ubiquitous, yet a clear gap remains between ordinary users and professional photographers, who can identify aesthetic issues and provide actionable shooting guidance during capture. We define this capability as aesthetic guidance (AG) -- an essential but largely underexplored domain in computational aesthetics. Existing multimodal large language models (MLLMs) primarily offer overly positive feedback, failing to identify issues or provide actionable guidance. Without AG capability, they cannot effectively identify distracting regions or optimize compositional balance, thus also struggling in aesthetic cropping, which aims to refine photo composition through reframing after capture. To address this, we introduce AesGuide, the first large-scale AG dataset and benchmark with 10,748 photos annotated with aesthetic scores, analyses, and guidance. Building upon it, we propose Venus, a two-stage framework that first empowers MLLMs with AG capability through progressively complex aesthetic questions and then activates their aesthetic cropping power via CoT-based rationales. Extensive experiments show that Venus substantially improves AG capability and achieves state-of-the-art (SOTA) performance in aesthetic cropping, enabling interpretable and interactive aesthetic refinement across both stages of photo creation. Code is available at https://github.com/PKU-ICST-MIPL/Venus_CVPR2026.

Problem

• aesthetic guidance
• multimodal large language models
• computational aesthetics
• aesthetic cropping
• photo composition

Innovation

• Aesthetic Guidance
• Multimodal Large Language Models
• Aesthetic Cropping
• Chain-of-Thought Reasoning
• Computational Aesthetics