Venus：面向美学指导与裁剪的多模态大语言模型基准构建与能力增强
Venus: Benchmarking and Empowering Multimodal Large Language Models for Aesthetic Guidance and Cropping
把「美学指导」定义为一个可评测的 MLLM 能力，并用两阶段（渐进式美学问答 + CoT 理由）微调把该能力迁移到裁剪重构图任务上，可作为条件化生成/编辑中「可解释美学先验」的可复用范式。
原文依据 we propose Venus, a two-stage framework that first empowers MLLMs with AG capability through progressively complex aesthetic questions
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论智能手机的普及使摄影变得无处不在，但普通用户与专业摄影师之间仍存在明显差距：后者能够在拍摄过程中识别美学问题并给出可操作的拍摄指导。我们将这一能力定义为美学指导（aesthetic guidance, AG）——它是计算美学中一个重要但长期被忽视的方向。现有多模态大语言模型（MLLM）主要给出过于正面的反馈，既无法识别问题，也无法提供可操作的指导。由于缺乏 AG 能力，它们无法有效识别干扰性区域或优化构图平衡，因而在旨在拍摄后通过重新取景来改善照片构图的美学裁剪任务上也表现不佳。为解决这一问题，我们提出了 AesGuide，这是首个大规模 AG 数据集与基准，包含 10,748 张照片，并标注了美学分数、分析与指导。在此基础上，我们提出 Venus，一个两阶段框架：首先通过逐步复杂化的美学问题增强 MLLM 的 AG 能力，随后借助基于思维链（CoT）的理由说明激活其美学裁剪能力。大量实验表明，Venus 显著提升了 AG 能力，并在美学裁剪上达到最优性能，从而在照片创作的两个阶段均实现可解释、可交互的美学优化。代码见 https://github.com/PKU-ICST-MIPL/Venus_CVPR2026。
英文原文摘要 1453 字符
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.