PulseAugur
实时 10:22:39
English(EN) TrAction: Action Recognition with Sparse Trajectories

TrAction 使用稀疏轨迹实现高效动作识别

研究人员开发了 TrAction,一种新颖的 Transformer 架构,用于使用稀疏点轨迹而非密集视频进行动作识别。该方法旨在减少依赖外观或背景线索的传统模型中存在的偏差。TrAction 在 Something-Something V2EPIC-Kitchens-100 等基准测试中取得了有竞争力的准确率,并且与其他模型融合后,性能得到进一步提升。 AI

影响 为视频动作识别提供了一种更高效、偏差更小的方法,有望提高现实世界应用中的性能。

排序理由 该集群包含一篇详细介绍新模型架构和实验结果的学术论文。

在 arXiv cs.CV 阅读 →

AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →

报道来源 [2]

  1. arXiv cs.CV TIER_1 English(EN) · Jan F. Meier, Felix B. Mueller, Alexander Ecker, Timo L\"uddecke ·

    TrAction: Action Recognition with Sparse Trajectories

    arXiv:2606.03490v1 Announce Type: new Abstract: Modern action recognition models operate on memory- and compute-intensive dense RGB video volumes and frequently exploit appearance and background shortcuts, for example, predicting actions from objects or scenes instead of characte…

  2. arXiv cs.CV TIER_1 English(EN) · Timo Lüddecke ·

    TrAction: Action Recognition with Sparse Trajectories

    Modern action recognition models operate on memory- and compute-intensive dense RGB video volumes and frequently exploit appearance and background shortcuts, for example, predicting actions from objects or scenes instead of characteristic motion. We investigate an efficient alter…