The Camera Inside the Editor: Reading the Implicit Camera of Image Editors with Painted Calibration Patterns
作者: Sebastian Rückerl
分类: cs.CV, cs.LG
发布日期: 2026-09-29
备注: 23 pages, 13 figures, 8 tables
💡 一句话要点
提出隐式相机模型以解析图像编辑器的绘制过程
🎯 匹配领域: 支柱三：空间感知与语义 (Perception & Semantics)
关键词: 隐式相机 图像编辑 相机校准 投影结构 几何参数
📋 核心要点
- 现有图像编辑器在绘制过程中假设的相机模型尚不明确，导致无法准确理解其绘制效果。
- 本文提出了一种通过绘制棋盘图案来推导隐式相机参数的方法，能够独立于训练数据进行相机识别。
- 实验结果显示，该方法在相机参数的匹配精度上优于传统校准工具GeoCalib，尤其在俯仰和焦距方面表现突出。
📝 摘要（中文）
基于指令的图像编辑器能够插入对象、重新设计场景并渲染新视角，但其在绘制照片时假设的相机模型尚不明确。本文通过绘制棋盘图案，利用投影结构推导出相机参数，包括俯仰、滚转、焦距和偏航等。与传统的相机校准工具GeoCalib不同，本文方法能够独立识别编辑器绘制时所使用的隐式相机。实验结果表明，Qwen-Image-Edit-2511在120个渲染相机上，绘制的瓷砖边缘与消失点的偏差仅为0.26度，隐式相机在俯仰和焦距上的匹配精度分别为0.8度和6%。此外，研究还发现隐式相机在绘制过程中表现出特定的先验知识，尤其是在真实照片上表现更为显著。
🔬 方法详解
问题定义：本文旨在解决图像编辑器在绘制过程中所假设的相机模型不明确的问题。现有方法如GeoCalib虽然可以估计图像的相机参数，但无法准确识别编辑器绘制时所用的隐式相机。
核心思路：论文通过让编辑器绘制棋盘图案，利用其投影结构来推导相机的俯仰、滚转、焦距和偏航等参数。这种方法不依赖于训练数据，能够直接从绘制结果中提取相机信息。
技术框架：整体流程包括：首先，编辑器根据指令绘制棋盘图案；其次，通过分析绘制的投影结构，提取相机的几何参数；最后，将提取的参数与真实相机参数进行对比，验证其准确性。
关键创新：最重要的创新点在于提出了一种新的隐式相机模型，通过绘制而非显式任务来识别相机参数。这一方法与传统的相机校准方法有本质区别，能够更好地适应图像编辑器的工作流程。
关键设计：在参数设置上，论文发现隐式相机的俯仰和焦距具有特定的先验知识，俯仰向水平面倾斜，焦距趋向于约30mm。此外，实验中还考虑了模糊对参数提取的影响，结果表明即使在模糊情况下，隐式相机的先验知识依然有效。
🖼️ 关键图片
📊 实验亮点
实验结果显示，Qwen-Image-Edit-2511在120个渲染相机上，绘制的瓷砖边缘与消失点的偏差仅为0.26度，隐式相机在俯仰和焦距上的匹配精度分别为0.8度和6%。与GeoCalib相比，本文方法在大多数情况下表现更为优越，尤其是在真实照片上。
🎯 应用场景
该研究的潜在应用领域包括图像编辑软件、虚拟现实和增强现实等场景。通过更准确地理解图像编辑器的隐式相机模型，开发者可以优化编辑工具，提高用户体验。此外，该技术也可用于自动化图像校准和增强图像合成的真实感。
📄 摘要（原文）
Instruction-based image editors insert objects, restyle scenes and render new viewpoints, but it is unknown which camera they assume when they paint into a photograph. Asked to cover the floor with a checkerboard, an editor paints projective structure from which classical vanishing-point geometry reads pitch, roll, focal length, yaw and, on renders, the principal point, without any training. Unlike a calibrator such as GeoCalib, which estimates the camera of an image, this isolates the camera under which the editor paints. On 120 rendered cameras with exact ground truth, Qwen-Image-Edit-2511 paints tile edges that meet their vanishing points within 0.26 degrees, and its implicit camera matches the true one to 0.8 degrees in pitch and 6% in focal length, more accurately than GeoCalib except in roll. Asked to draw the horizon or mark a vanishing point instead, the editor fails, so this knowledge is revealed by painting and not by the explicit tasks we tried. The implicit camera has two priors: roll is pulled towards level (slope 0.71), and telephoto perspective towards a default of about 30 mm, which roughly matches the camera the models paint without any scene. For Qwen, the priors do not grow when blur removes four fifths of the line evidence. They are stronger on real photographs, and on NYUv2 a shorter wording of the task removes the difference for roll. On photographs from a 24--240 mm zoom lens the painted perspective grows with only 0.62 of the lens's slope, while GeoCalib and MoGe-2 saturate at about 52 and 42 mm. FLUX.1 Kontext and LongCat-Image-Edit are pulled much harder. Finally, from a level camera a camera-control LoRA executes pose commands at only 50--70% of their strength, and a board painted into its output agrees with the camera it produced.