Aesthetic Camera Viewpoint Suggestion with 3D Aesthetic Field

TL;DR

This paper introduces a 3D aesthetic field model that enables efficient, geometry-grounded camera viewpoint suggestions using sparse data, outperforming existing methods that rely on dense captures or reinforcement learning.

Contribution

We propose a novel 3D aesthetic field learned via a Gaussian Splatting network, enabling aesthetic prediction from sparse views and a two-stage search for optimal viewpoints.

Findings

Outperforms existing viewpoint suggestion methods in aesthetic quality

Efficiently predicts viewpoints using sparse input views

Establishes a new direction for 3D-aware aesthetic modeling

Abstract

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.