TGT：面向局部可控视频生成的文本接地轨迹
TGT: Text-Grounded Trajectories for Locally Controlled Video Generation
把点轨迹当作直观的运动句柄，并与局部文本配对，通过位置感知交叉注意力注入扩散/流模型，为视频生成提供细粒度、多目标的局部运动与外观控制；双CFG分别调节局部与全局文本引导的思路可直接迁移到其他条件控制生成任务。
原文依据 We propose Location-Aware Cross-Attention (LACA) to integrate these signals and adopt a dual-CFG scheme to separately modulate local and global text guidance.
摘要译文
由模型翻译 · 术语统一 · 请对照原文核对关键结论文本到视频生成在视觉保真度方面已快速进步，而标准方法对生成场景中主体构成的控制能力仍然有限。已有工作表明，加入边界框或分割掩码等局部化文本控制信号会有所帮助。然而，这些方法在复杂场景中表现不佳，在多目标设定下性能下降，精度有限，并且随着可控目标数量增加，单条轨迹与视觉实体之间缺乏清晰的对应关系。我们提出文本接地轨迹（TGT），一个以轨迹配以局部化文本描述为条件来生成视频的框架。我们提出位置感知交叉注意力（LACA）来整合这些信号，并采用双CFG方案分别调制局部与全局文本引导。此外，我们开发了一条数据处理流程，用于生成带有被跟踪实体局部描述的轨迹，并标注了两百万段高质量视频片段以训练TGT。这些组件共同使TGT能够将点轨迹用作直观的运动句柄，为每条轨迹配对文本，从而同时控制外观与运动。大量实验表明，与已有方法相比，TGT 取得了更高的视觉质量、更准确的文本对齐以及更好的运动可控性。
英文原文摘要 1378 字符
Text-to-video generation has advanced rapidly in visual fidelity, whereas standard methods still have limited ability to control the subject composition of generated scenes. Prior work shows that adding localized text control signals, such as bounding boxes or segmentation masks, can help. However, these methods struggle in complex scenarios and degrade in multi-object settings, offering limited precision and lacking a clear correspondence between individual trajectories and visual entities as the number of controllable objects increases. We introduce Text-Grounded Trajectories (TGT), a framework that conditions video generation on trajectories paired with localized text descriptions. We propose Location-Aware Cross-Attention (LACA) to integrate these signals and adopt a dual-CFG scheme to separately modulate local and global text guidance. In addition, we develop a data processing pipeline that produces trajectories with localized descriptions of tracked entities, and we annotate two million high quality video clips to train TGT. Together, these components enable TGT to use point trajectories as intuitive motion handles, pairing each trajectory with text to control both appearance and motion. Extensive experiments show that TGT achieves higher visual quality, more accurate text alignment, and improved motion controllability compared with prior approaches.