DiffBMP：基于位图图元的可微渲染
DiffBMP: Differentiable Rendering with Bitmap Primitives
提供了一种可微的位图图元渲染引擎，视觉研究者可将其用于图像/视频的图层分解、基于位图的场景表示优化，或作为可微渲染管线中处理位图元素的替代方案。
原文依据 a scalable and efficient differentiable rendering engine for a collection of bitmap images
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论我们介绍了 DiffBMP，一个面向位图图像集合的可扩展且高效的可微渲染引擎。我们的工作解决了一个限制：传统的可微渲染器受限于矢量图形，而世界上大多数图像是位图。我们的核心贡献是一个高度并行的渲染管线，其特点是用于计算梯度的自定义 CUDA 实现。该系统例如可以在消费级 GPU 上于 1 分钟内优化数千个位图图元的位置、旋转、缩放、颜色和不透明度。我们采用并验证了若干技术以促进优化：通过高斯模糊的软光栅化、结构感知初始化、噪声画布，以及针对视频或空间受限图像的专用损失/启发式方法。我们证明 DiffBMP 不仅是一个孤立的工具，而且是一个旨在集成到创意工作流中的实用工具。它支持将合成结果导出为原生分层文件格式，并且整个框架可通过易于修改的 Python 包公开访问。
英文原文摘要 1082 字符
We introduce **DiffBMP**, a scalable and efficient differentiable rendering engine for a collection of bitmap images. Our work addresses a limitation that traditional differentiable renderers are constrained to vector graphics, given that most images in the world are bitmaps. Our core contribution is a highly parallelized rendering pipeline, featuring a custom CUDA implementation for calculating gradients. This system can, for example, optimize the position, rotation, scale, color, and opacity of thousands of bitmap primitives all in under 1 min using a consumer GPU. We employ and validate several techniques to facilitate the optimization: soft rasterization via Gaussian blur, structure-aware initialization, noisy canvas, and specialized losses/heuristics for videos or spatially constrained images. We demonstrate DiffBMP is not just an isolated tool, but a practical one designed to integrate into creative workflows. It supports exporting compositions to a native, layered file format, and the entire framework is publicly accessible via an easy-to-hack Python package.