基于三维美学场的相机视角美学建议
Aesthetic Camera Viewpoint Suggestion with 3D Aesthetic Field
用前馈 3D 高斯泼溅网络把预训练 2D 美学模型的打分蒸馏成三维美学场，使仅凭稀疏视角即可对新视角做美学预测；这套「2D 先验蒸馏进 3D 表示 + 粗采样加梯度精修」的流程，可迁移到任何需要在三维空间中优化相机或视角的视觉任务。
原文依据 learn this 3D aesthetic field using a feedforward 3D Gaussian Splatting network that distills high-level aesthetic knowledge from a pretrained 2D aesthetic model into 3D space
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论场景的美学质量在很大程度上取决于相机视角。现有的美学视角建议方法要么是单视图调整，即从单张图像预测有限的相机调整量而不理解场景几何；要么是三维探索方法，依赖稠密采集或预先构建的三维环境，并配合代价高昂的强化学习搜索。在本工作中，我们引入三维美学场这一概念，使得在稀疏采集条件下即可进行几何接地的三维美学推理，从而相比代价高昂的强化学习搜索能够高效地给出视角建议。我们选择使用前馈三维高斯泼溅网络来学习该三维美学场，将预训练二维美学模型中的高层美学知识蒸馏到三维空间，从而仅凭稀疏输入视图即可对新视角进行美学预测。基于该场，我们提出一个两阶段搜索流程，将粗粒度视角采样与基于梯度的精修相结合，无需稠密采集或强化学习探索即可高效地找出具有美学吸引力的视角。大量实验表明，与现有方法相比，我们的方法能够稳定地给出构图与取景更优的视角建议，为三维感知的美学建模开辟了新方向。
英文原文摘要 1333 字符
The aesthetic quality of a scene depends strongly on camera viewpoint. Existing approaches for aesthetic viewpoint suggestion are either single-view adjustments, predicting limited camera adjustments from a single image without understanding scene geometry, or 3D exploration approaches, which rely on dense captures or prebuilt 3D environments coupled with costly reinforcement learning (RL) searches. In this work, we introduce the notion of 3D aesthetic field that enables geometry-grounded aesthetic reasoning in 3D with sparse captures, allowing efficient viewpoint suggestions in contrast to costly RL searches. We opt to learn this 3D aesthetic field using a feedforward 3D Gaussian Splatting network that distills high-level aesthetic knowledge from a pretrained 2D aesthetic model into 3D space, enabling aesthetic prediction for novel viewpoints from only sparse input views. Building on this field, we propose a two-stage search pipeline that combines coarse viewpoint sampling with gradient-based refinement, efficiently identifying aesthetically appealing viewpoints without dense captures or RL exploration. Extensive experiments show that our method consistently suggests viewpoints with superior framing and composition compared to existing approaches, establishing a new direction toward 3D-aware aesthetic modeling.