arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.13141cs.CV
收藏

用于时态动作建议细化的时态上下文聚合网络

Temporal Context Aggregation Network for Temporal Action Proposal Refinement

Huazhong University of Science and Technology(华中科技大学);SenseTime Research(商汤科技研究院);Shanghai AI Laboratory(上海人工智能实验室)

Zhiwu Qing, Haisheng Su, Weihao Gan, Dongliang Wang, Wei Wu, Xiang Wang, Yu Qiao, Junjie Yan, Changxin Gao, Nong Sang

摘要:

时间动作建议生成的目的是估计未修剪视频中动作的时间间隔,这是视频理解领域中具有挑战性且重要的任务。由于缺乏有效的时间建模和有效的边界上下文利用,目前的方法产生的建议仍然遭受不准确的时间边界和用于检索的置信度较低。在本文中,我们提出了时间上下文聚合网络(TCANet)生成高质量的行动建议,通过“本地和全球”的时间上下文聚合和补充,以及渐进的边界细化。具体来说,我们首先设计了一个局部-全局时间编码器(LGTE),它采用了信道分组策略,有效地编码“局部和全局”的时间相互依赖。此外,建议的边界和内部上下文分别用于帧级和段级边界回归。时间边界回归器(TBR)旨在以端到端的方式将这两个回归粒度结合起来,通过渐进式细化实现精确的边界和可靠的建议置信度。在三个具有挑战性的数据集上进行了广泛的实验:HACS,ActivityNet-v1.3和THUMOS-14,其中TCANet可以生成具有高精度和召回率的建议。通过与现有的动作分类器相结合,TCANet可以获得显着的时间动作检测性能相比,其他方法。毫不奇怪,提出的TCANet在CVPR 2020 - HACS挑战排行榜上赢得了时间动作本地化任务的1个$^{st}$位置。

英文摘要:

Temporal action proposal generation aims to estimate temporal intervals of actions in untrimmed videos, which is a challenging yet important task in the video understanding field. The proposals generated by current methods still suffer from inaccurate temporal boundaries and inferior confidence used for retrieval owing to the lack of efficient temporal modeling and effective boundary context utilization. In this paper, we propose Temporal Context Aggregation Network (TCANet) to generate high-quality action proposals through "local and global" temporal context aggregation and complementary as well as progressive boundary refinement. Specifically, we first design a Local-Global Temporal Encoder (LGTE), which adopts the channel grouping strategy to efficiently encode both "local and global" temporal inter-dependencies. Furthermore, both the boundary and internal context of proposals are adopted for frame-level and segment-level boundary regressions, respectively. Temporal Boundary Regressor (TBR) is designed to combine these two regression granularities in an end-to-end fashion, which achieves the precise boundaries and reliable confidence of proposals through progressive refinement. Extensive experiments are conducted on three challenging datasets: HACS, ActivityNet-v1.3, and THUMOS-14, where TCANet can generate proposals with high precision and recall. By combining with the existing action classifier, TCANet can obtain remarkable temporal action detection performance compared with other methods. Not surprisingly, the proposed TCANet won the 1$^{st}$ place in the CVPR 2020 - HACS challenge leaderboard on temporal action localization task.

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。

↑