用几何路径匹配提升新视角生成的一致性
GeodesicNVS: Probability Density Geodesic Flow Matching for Novel View Synthesis
- 构建数据到数据的确定性变换框架，显式耦合视图关系
- 在Objaverse和GSO30上超越扩散模型，视图过渡更平滑
- 适合关注视图一致性与几何结构建模的研究者
近期生成模型进展显著提升了新视角合成效果，但跨视角一致性仍是难题。基于扩散的模型依赖随机噪声到数据的转换，掩盖了确定性结构，导致视图预测不一致。本文提出数据到数据的流匹配框架，学习成对视图间的确定性变换，通过显式数据耦合提升视图一致性。在此基础上，提出概率密度测地线流匹配（PDG-FM），使插值轨迹与数据流形上的密度测地线对齐。为实现可计算的测地线估计，采用教师-学生框架，将密度依赖的测地线插值器压缩为高效的环境空间预测器。实验表明，该方法在Objaverse和GSO30数据集上优于扩散基线，显著提升结构连贯性与视图间过渡平滑度。结果验证了将数据相关几何正则化引入确定性流匹配，在一致新视角生成中的优势。
原文摘要 · Abstract (English)
Recent advances in generative modeling have substantially enhanced novel view synthesis, yet maintaining consistency across viewpoints remains challenging. Diffusion-based models rely on stochastic noise-to-data transitions, which obscure deterministic structures and yield inconsistent view predictions. We advocate a Data-to-Data Flow Matching framework that learns deterministic transformations between paired views, enhancing view-consistent synthesis through explicit data coupling. Building on this, we propose Probability Density Geodesic Flow Matching (PDG-FM), which aligns interpolation trajectories with density-based geodesics of a data manifold. To enable tractable geodesic estimation, we employ a teacher-student framework that distills density-based geodesic interpolants into an efficient ambient-space predictor. Empirically, our method surpasses diffusion-based baselines on Objaverse and GSO30 datasets, demonstrating improved structural coherence and smoother transitions across views. These results highlight the advantages of incorporating data-dependent geometric regularization into deterministic flow matching for consistent novel view generation.
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。