arXiv 2603.226062026-03-23cs.CV

TrajLoom：从视频生成密集未来轨迹

TrajLoom: Dense Future Trajectory Generation from Video

Zewei Zhang, Jia Jun Cheng Xian, Kaiwen Liu, Ming Liang, Hang Chu, Jun Chen, Renjie Liao

备注：: Project page, code, model checkpoints, and datasets: https://trajloom.github.io/

AI总结：

本文提出TrajLoom框架，通过过去轨迹和视频上下文预测未来轨迹和可见性，采用Grid-Anchor Offset Encoding、TrajLoom-VAE和TrajLoom-Flow三种组件，扩展预测时间范围并提升轨迹真实性和稳定性。

摘要：

预测未来运动对于视频理解和可控视频生成至关重要。密集点轨迹是紧凑且表达性强的运动表示，但从观察视频中建模其未来演变仍具挑战性。我们提出一个框架，通过过去轨迹和视频上下文预测未来轨迹和可见性。我们的方法包含三个组件：（1）Grid-Anchor Offset Encoding，通过将每个点表示为像素中心锚点的偏移量来减少位置依赖偏差；（2）TrajLoom-VAE，通过掩码重建和时空一致性正则化学习紧凑的时空潜在空间；（3）TrajLoom-Flow，通过流匹配在潜在空间中生成未来轨迹，结合边界提示和on-policy K-step微调以实现稳定采样。我们还引入TrajLoomBench，一个涵盖真实和合成视频的统一基准，其标准化设置与视频生成基准对齐。与最先进的方法相比，我们的方法将预测时间范围从24帧扩展到81帧，同时在多个数据集上提升运动真实性和稳定性。预测轨迹直接支持下游视频生成和编辑。代码、模型检查点和数据集可在https://trajloom.github.io/获取。

英文摘要：

Predicting future motion is crucial in video understanding and controllable video generation. Dense point trajectories are a compact, expressive motion representation, but modeling their future evolution from observed video remains challenging. We propose a framework that predicts future trajectories and visibility from past trajectories and video context. Our method has three components: (1) Grid-Anchor Offset Encoding, which reduces location-dependent bias by representing each point as an offset from its pixel-center anchor; (2) TrajLoom-VAE, which learns a compact spatiotemporal latent space for dense trajectories with masked reconstruction and a spatiotemporal consistency regularizer; and (3) TrajLoom-Flow, which generates future trajectories in latent space via flow matching, with boundary cues and on-policy K-step fine-tuning for stable sampling. We also introduce TrajLoomBench, a unified benchmark spanning real and synthetic videos with a standardized setup aligned with video-generation benchmarks. Compared with state-of-the-art methods, our approach extends the prediction horizon from 24 to 81 frames while improving motion realism and stability across datasets. The predicted trajectories directly support downstream video generation and editing. Code, model checkpoints, and datasets are available at https://trajloom.github.io/.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。