ASTRA：基于轨迹对齐的异步时空重建
ASTRA: Asynchronous Spatio-Temporal Reconstruction via Trajectory Alignment
浏览论文内容
中文总结 AI 辅助
针对动态三维场景重建中多相机异步导致的问题，提出ASTRA框架，通过轨迹对齐联合优化时间偏移与三维表示，在实验中实现了显著的性能提升。
中文摘要 AI 辅助
动态三维场景重建在严格同步的多相机输入假设下已取得显著成功，但在真实场景中，采集设备间的时间异步仍是关键挑战，会导致严重的运动模糊和几何伪影。现有异步重建方法通常通过光度监督估计时间偏移，但在大偏移和复杂运动下，外观匹配提供的时间线索很弱。我们将此局限归因于两个主要瓶颈：纹理诱导崩溃，即低纹理区域提供几乎消失的对齐信号；变形诱导耦合，即时间误差被吸收到失真的几何或运动中，而非被显式校正。为解决这些问题，我们提出ASTRA（Asynchronous Spatio-Temporal Reconstruction via Trajectory Alignment），该框架引入二维运动轨迹作为异步动态重建的显式、与纹理无关的监督。ASTRA不再仅通过渲染的颜色残差同步相机，而是通过将重建三维点的投影运动与观测到的二维轨迹对齐，联合优化时间偏移和动态三维表示，同时使用动态和确定性掩码抑制不可靠的轨迹约束。在不同动态高斯溅射（Dynamic Gaussian Splatting）主干上的大量实验表明，ASTRA能保留高频空间细节，即使在高达25帧偏移的严重异步情况下仍保持强鲁棒性，实现约1.4 dB的峰值信噪比（PSNR）提升，将时间偏移的平均绝对误差（MAE）降低54.0%，同步成功率几乎提高三倍。
英文摘要
Dynamic 3D scene reconstruction has made significant progress with multi-camera systems, often relying on temporally aligned observations across views. However, in real-world scenarios, temporal asynchrony among capturing devices remains a common limitation, leading to severe motion blur and geometric artifacts. Existing asynchronous reconstruction methods typically estimate temporal offsets through photometric supervision, but appearance matching provides weak temporal cues under large offsets and complex motions. We attribute this limitation to two critical issues: texture-induced collapse, where low-textured regions provide nearly vanishing alignment signals, and deformation-induced entanglement, where temporal errors are absorbed into distorted geometry or motion rather than being explicitly corrected. To address these issues, we propose ASTRA (Asynchronous Spatio-Temporal Reconstruction via Trajectory Alignment), a framework that introduces 2D motion trajectories as explicit, texture-robust supervision for asynchronous dynamic reconstruction. Instead of synchronizing cameras solely through rendered color residuals, ASTRA jointly optimizes temporal offsets and dynamic 3D representations by aligning the projected motion of reconstructed 3D points with observed 2D trajectories, while using dynamic and certainty masking to suppress unreliable trajectory constraints. Extensive experiments on different dynamic Gaussian Splatting backbones show that ASTRA preserves high-frequency spatial details and sustains strong robustness even under severe asynchrony with up to 25-frame offsets, achieving approximately 1.4 dB PSNR improvement, reducing temporal-offset MAE by 54.0%, and nearly quadrupling the synchronization success rate.
发表机构
- School of Electronic Science and Engineering, Nanjing University（南京大学电子科学与工程学院）
机构由 AI 辅助整理，请以论文原文为准。
