CubicSplat：通过误差有界前向松弛实现可微矢量图形
CubicSplat: Differentiable Vector Graphics via Error-Bounded Forward Relaxation
- State Key Laboratory of Cognitive Intelligence, University of Science and Technology of China（中国科学技术大学认知智能国家重点实验室）
- Hefei Normal University（合肥师范学院）
- Zhejiang Normal University（浙江师范学院）
机构由 AI 辅助整理，请以论文原文为准。
AI总结：
该研究针对矢量图形可微优化的梯度矛盾问题，提出CubicSplat可微矢量光栅化器，在DIV2K和Kodak基准上实现更优重建质量与更快训练速度
AI中文摘要：
矢量图形因具备分辨率独立性、存储紧凑性和直接可编辑性而备受青睐，这使得对其参数化基元进行可微优化成为极具吸引力的目标。然而，经典光栅化在几何层面是不连续的，现有用于平滑前向传播的补救措施随着场景复杂度的提升需要愈发复杂的启发式方法。我们将这种脆弱性归因于梯度跷跷板：提升前向几何精确性的设计选择会系统性地劣化诱导的梯度信号，反之亦然。为应对这一矛盾，我们提出了CubicSplat，这是一种可微矢量光栅化器，它用几何误差为$O(S^{-2})$的均匀折线替代Bézier最近点求解器。由此产生的静态计算图从结构上产生了条件良好的梯度，同时一种源自合成的可见性机制无需辅助正则化即可修剪退化基元。在DIV2K和Kodak基准测试中，CubicSplat在闭合填充设置下实现了最先进的重建质量，PSNR提升超过2 dB，同时训练速度比现有方法快4倍以上。代码可在this https URL获取
英文摘要：
Vector graphics are prized for their resolution independence, compact storage, and direct editability, making differentiable optimization of their parametric primitives an attractive goal. Yet classical rasterization is discontinuous with respect to geometry, and existing remedies that smooth the forward pass demand increasingly elaborate heuristics as scene complexity grows. We trace this fragility to a gradient seesaw: design choices that improve forward geometric exactness can systematically degrade the induced gradient signal, and vice versa. To navigate this tension we introduce CubicSplat, a differentiable vector rasterizer that replaces Bézier closest-point solvers with uniform polyline surrogates whose geometric error is bounded at $O(S^{-2})$. The resulting static computation graph yields well-conditioned gradients by construction, while a compositing-derived visibility mechanism prunes degenerate primitives without auxiliary regularization. On DIV2K and Kodak benchmarks CubicSplat achieves state-of-the-art reconstruction quality with over 2 dB PSNR gain in the closed-fill setting, while training up to 4x faster than prior methods. The code is available at https://github.com/CubicSplat/repo
