Sim-DETR: 解锁用于时间句子定位的DETR
Sim-DETR: Unlock DETR for Temporal Sentence Grounding
AI总结：
Sim-DETR通过改进DETR的解码器层，解决了时间句子定位中的查询冲突问题，提升了模型性能。
摘要：
时间句子定位旨在识别视频中与给定文本查询对应的精确时刻，通常采用检测转换器（DETR）解决方案。然而，我们发现典型的增强DETR的策略并未提高，甚至可能降低其在该任务中的性能。我们系统地分析并识别了这种异常行为的根本原因：（1）相似目标时刻的查询之间的冲突；（2）由于全局语义与局部定位之间的张力导致的内部查询冲突。基于这些见解，我们提出了一种简单而强大的基线，Sim-DETR，它通过在解码器层中进行两个微小修改来扩展标准DETR：（1）根据查询的语义和位置重叠约束自我注意力；（2）添加查询到帧对齐，以连接全局和局部上下文。实验表明，Sim-DETR解锁了DETR在时间句子定位中的全部潜力，为未来研究提供了强有力的基线。
英文摘要：
Temporal sentence grounding aims to identify exact moments in a video that correspond to a given textual query, typically addressed with detection transformer (DETR) solutions. However, we find that typical strategies designed to enhance DETR do not improve, and may even degrade, its performance in this task. We systematically analyze and identify the root causes of this abnormal behavior: (1) conflicts between queries from similar target moments and (2) internal query conflicts due to the tension between global semantics and local localization. Building on these insights, we propose a simple yet powerful baseline, Sim-DETR, which extends the standard DETR with two minor modifications in the decoder layers: (1) constraining self-attention between queries based on their semantic and positional overlap and (2) adding query-to-frame alignment to bridge the global and local contexts. Experiments demonstrate that Sim-DETR unlocks the full potential of DETR for temporal sentence grounding, offering a strong baseline for future research.
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。
