离散标注，连续偏好：重新思考美学图像裁剪的监督方式以实现准确与泛化
Discrete Annotation, Continuous Preference: Rethinking Supervision for Accurate and Generalizable Aesthetic Image Cropping
浏览论文内容
中文总结 AI 辅助
针对美学图像裁剪中标注主观性和离散性问题，提出连续偏好场（CPF）建模偏好，训练VLM模型CPIC并校准基准CPICD，实现最先进性能和泛化。
中文摘要 AI 辅助
美学图像裁剪旨在根据美学和构图识别图像的最佳裁剪区域。尽管基于标注数据的监督是基础，但该领域长期受到一个问题的阻碍：现有数据集存在（1）人类主观性和（2）局限于固定采样网格的刚性离散性。这些有缺陷的标注不仅限制了训练模型的准确性和泛化能力，还严重扭曲了公平评估。为克服这一问题，我们提出将人类裁剪偏好建模为裁剪空间上的一个多峰、连续且尖锐的场。我们引入了连续偏好场（CPF），通过（1）峰值聚类、（2）离格细化、（3）负向塑造和（4）场组装，从离散标注中恢复密集的偏好景观。基于此，我们训练了CPIC，一个基于VLM的裁剪模型，通过GRPO和CPF奖励进行优化，克服了模板坍缩，实现了最先进的性能和卓越的域外泛化。最后，为解决长期存在的基准评估危机，我们引入了CPICD，对现有真实边界框进行全面重新校准。通过利用CPF纠正主流基准中的网格边界伪影，CPICD为未来的裁剪研究建立了严谨可靠的基础。大量实验和用户研究证明了我们的CPF、CPIC和CPICD的优越性。代码、模型和数据可在该https URL获取。
英文摘要
Aesthetic image cropping aims to identify the optimal crop of an image in terms of aesthetics and composition. While supervision based on annotated data is fundamental, the field has been hindered by a long-standing problem: existing datasets suffer from (1) human subjectivity and (2) rigid discreteness confined to fixed sampling grids. These flawed annotations not only limit the accuracy and generalization of trained models but also severely distort fair evaluation. To overcome this, we propose to model human cropping preference as a multi-peaked, continuous, and sharp field over the crop space. We introduce the Continuous Preference Field (CPF), which recovers a dense preference landscape from discrete annotations through (1) peak clustering, (2) off-lattice refinement, (3) negative shaping, and (4) field assembly. Based on this, we train CPIC, a VLM-based cropping model optimized via GRPO with the CPF reward, which overcomes template collapse, achieving state-of-the-art performance and exceptional out-of-domain generalization. Finally, to resolve the long-standing benchmark evaluation crisis, we introduce CPICD, a comprehensive recalibration of existing ground-truth boxes. By leveraging the CPF to correct grid-bound artifacts across mainstream benchmarks, CPICD establishes a rigorous and reliable foundation for future cropping research. Extensive experiments and user studies demonstrate the superiority of our CPF, CPIC, and CPICD. Code, model, and data are available at https://github.com/zzqingz/CPIC.
发表机构
- Shanghai Jiao Tong University（上海交通大学）
- Huawei（华为）
机构由 AI 辅助整理，请以论文原文为准。
