arXiv 2006.075262020-06-24cs.CV

CBR-Net：用于行动检测的级联边界细化网络：提交ActivityNet 2020年挑战赛（任务1）

CBR-Net: Cascade Boundary Refinement Network for Action Detection: Submission to ActivityNet Challenge 2020 (Task 1)

Huazhong University of Science and Technology（华中科技大学）；DAMO Academy, Alibaba Group（阿里巴巴集团达摩院）

Xiang Wang, Baiteng Ma, Zhiwu Qing, Yongpeng Sang, Changxin Gao, Shiwei Zhang, Nong Sang

备注：: ActivityNet Challenge 2020 Temporal Action Localization (Task 1) Champion Solution (Rank 1)

摘要：

在本报告中，我们提出了我们在ActivityNet Challenge 2020中的时间动作定位（检测）任务（任务1）的解决方案。该任务的目的是在时间上定位感兴趣的动作发生的时间间隔，并预测长的未修剪视频中的动作类别。我们的解决方案主要包括三个部分：1）特征编码：我们应用了三种骨干，包括TSN [7]，Slowfast[3]和I3 d [1]，它们都是在Kinetics数据集上预训练的[2]。应用这些模型，我们可以提取片段级别的视频表示; 2）提案生成：我们选择BMN [5]作为我们的基线，在此基础上，我们设计了级联边界细化网络（CBR-Net）来进行提案检测。CBR-Net主要包含两个模块：时间特征编码，应用BiLSTM对长期时间信息进行编码; CBR模块，旨在提高不同参数设置下的提案精度; 3）动作定位：在这个阶段，我们结合微调网络获得的视频级分类结果来预测每个提案的类别。此外，我们还应用不同的集成策略来提高所设计的解决方案的性能，通过这些策略，我们在ActivityNet v1.3数据集的测试集上实现了42.788%的平均平均精度指标。

英文摘要：

In this report, we present our solution for the task of temporal action localization (detection) (task 1) in ActivityNet Challenge 2020. The purpose of this task is to temporally localize intervals where actions of interest occur and predict the action categories in a long untrimmed video. Our solution mainly includes three components: 1) feature encoding: we apply three kinds of backbones, including TSN [7], Slowfast[3] and I3d[1], which are both pretrained on Kinetics dataset[2]. Applying these models, we can extract snippet-level video representations; 2) proposal generation: we choose BMN [5] as our baseline, base on which we design a Cascade Boundary Refinement Network (CBR-Net) to conduct proposal detection. The CBR-Net mainly contains two modules: temporal feature encoding, which applies BiLSTM to encode long-term temporal information; CBR module, which targets to refine the proposal precision under different parameter settings; 3) action localization: In this stage, we combine the video-level classification results obtained by the fine tuning networks to predict the category of each proposal. Moreover, we also apply to different ensemble strategies to improve the performance of the designed solution, by which we achieve 42.788% on the testing set of ActivityNet v1.3 dataset in terms of mean Average Precision metrics.

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。

摘要：

英文摘要：

相关论文