用自适应小波表示动态场景，兼顾细节与运动连续性。
AdaGaR: Adaptive Gabor Representation for Dynamic Scene Reconstruction
- 用可学习频率权重和能量补偿的自适应小波替代高斯点
- 在Tap-Vid DAVIS上达到PSNR 35.49、SSIM 0.9433的领先效果
- 适合做视频插值、深度一致性、编辑和立体视图合成
从单目视频重建动态3D场景需同时捕捉高频外观细节与时间连续运动。现有基于单个高斯原型的方法受限于低通滤波特性，而标准小波函数易引发能量不稳。此外，缺乏时间连续性约束常导致插值时出现运动伪影。本文提出AdaGaR，统一解决显式动态建模中的频率自适应与时间连续性问题。引入自适应小波表示，通过可学习频率权重与自适应能量补偿，平衡细节捕获与稳定性。为保障时间连续性，采用带时间曲率正则化的三次埃尔米特样条，确保运动演化平滑。设计自适应初始化机制，结合深度估计、点追踪与前景掩码，在训练初期建立稳定的点云分布。在Tap-Vid DAVIS数据集上的实验表明，该方法达到当前最优性能（PSNR 35.49，SSIM 0.9433，LPIPS 0.0723），并在视频插值、深度一致性、视频编辑和立体视图合成任务中展现出强泛化能力。
原文摘要 · Abstract (English)
Reconstructing dynamic 3D scenes from monocular videos requires simultaneously capturing high-frequency appearance details and temporally continuous motion. Existing methods using single Gaussian primitives are limited by their low-pass filtering nature, while standard Gabor functions introduce energy instability. Moreover, lack of temporal continuity constraints often leads to motion artifacts during interpolation. We propose AdaGaR, a unified framework addressing both frequency adaptivity and temporal continuity in explicit dynamic scene modeling. We introduce Adaptive Gabor Representation, extending Gaussians through learnable frequency weights and adaptive energy compensation to balance detail capture and stability. For temporal continuity, we employ Cubic Hermite Splines with Temporal Curvature Regularization to ensure smooth motion evolution. An Adaptive Initialization mechanism combining depth estimation, point tracking, and foreground masks establishes stable point cloud distributions in early training. Experiments on Tap-Vid DAVIS demonstrate state-of-the-art performance (PSNR 35.49, SSIM 0.9433, LPIPS 0.0723) and strong generalization across frame interpolation, depth consistency, video editing, and stereo view synthesis. Project page: https://jiewenchan.github.io/AdaGaR/
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。