PhotoFlow: Agentic 3D Virtual Photography Missions

2026-05-22

AI Summary

This work addresses the challenge of generating high-quality virtual photographs in 3D scenes without prior camera poses or reference images, guided solely by natural language instructions, while ensuring both aesthetic appeal and spatial plausibility. To this end, the authors propose PhotoFlow, a closed-loop agent framework comprising director, critic, and reflection modules, which formulates language-guided virtual photography as an executable agent-based task for the first time. The approach integrates soft photographic blueprint generation, multi-stage camera optimization, region-aware memory, and dead-zone suppression, complemented by a vision-language joint evaluation strategy. The authors also introduce VPhotoBench, a new benchmark encompassing 47 scenes and 141 tasks. Experimental results demonstrate that under a six-rendering budget, PhotoFlow significantly outperforms existing baselines in terms of external quality alignment and task success rate.

Abstract

Virtual photography asks an agent to enter a prepared 3D scene with no preselected camera pose or reference image, infer a suitable shot from scene information and a language intent, choose executable camera parameters, and render the final photograph. Recent progress in vision-language models makes this kind of spatial agent increasingly plausible, but the task stresses two capabilities that remain hard to evaluate together: complex 3D spatial understanding and abstract aesthetic judgment. We introduce PhotoFlow, a Director-Reviewer-Reflector agent for closed-loop camera search. The Director builds a soft photographic blueprint and proposes diverse candidate cameras; the Reviewer combines rule checks, visual critique, and pairwise incumbent selection; and the Reflector converts failures into region memory, dead-zone suppression, and high-explore relocation. We also introduce VPhotoBench, a benchmark of 47 open-license Blender scenes and 141 language-conditioned photography missions spanning subject placement, relational composition, and atmosphere/style. On held-out experiments, PhotoFlow achieves the strongest external quality-alignment composite and success rate among one-shot prediction, single-chain reflection, anchor-bank selection, and random search under a six-round rendering budget. To our knowledge, this is the first work to make language-conditioned virtual photography in arbitrary Blender scenes an executable agent task, and our results show that an LLM-centered spatial agent can already produce strong photographs in a setting designed to challenge both 3D reasoning and aesthetic choice.

Problem

• virtual photography
• 3D spatial understanding
• aesthetic judgment
• language-conditioned generation
• camera pose selection

Innovation

• virtual photography
• spatial agent
• language-conditioned rendering
• 3D scene understanding
• aesthetic reasoning