用中间状态序列替代初始噪声，降低内存消耗并提升图像重建质量
Trajectory Stitching for Solving Inverse Problems with Flow-Based Models
- 将生成轨迹拆分为多个中间隐状态，避免全程反向传播
- 在修复、超分辨和断层扫描任务中均优于现有方法
- 适合需要高效高质重建的逆问题场景
基于流的生成模型已成为求解逆问题的强大先验。一种方法是直接优化初始隐变量（噪声），使流输出满足逆问题要求，但需对整个生成轨迹进行反向传播，导致内存开销大且数值不稳定。本文提出MS-Flow，将轨迹表示为一系列中间隐状态而非单一初始编码。通过局部施加流动力学约束，并利用轨迹匹配惩罚耦合各段，实现中间隐状态更新与观测数据一致性的交替优化。该方法显著降低内存消耗，同时提升重建质量。我们在图像恢复及各类逆问题（包括补全、超分辨、计算机断层成像）上验证了MS-Flow的有效性，结果优于现有方法。
原文摘要 · Abstract (English)
Flow-based generative models have emerged as powerful priors for solving inverse problems. One option is to directly optimize the initial latent code (noise), such that the flow output solves the inverse problem. However, this requires backpropagating through the entire generative trajectory, incurring high memory costs and numerical instability. We propose MS-Flow, which represents the trajectory as a sequence of intermediate latent states rather than a single initial code. By enforcing the flow dynamics locally and coupling segments through trajectory-matching penalties, MS-Flow alternates between updating intermediate latent states and enforcing consistency with observed data. This reduces memory consumption while improving reconstruction quality. We demonstrate the effectiveness of MS-Flow over existing methods on image recovery and inverse problems, including inpainting, super-resolution, and computed tomography.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。