.png){kind=icon}
返回
Compact Representation and Reliable Classification Learning for Point-Level Weakly-Supervised Action Localization
.png){kind=icon}
PRE
AI
DOI：10.1109/TIP.2022.3222623.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Point-level weakly-supervised temporal action localization (P-WSTAL) aims to localize temporal extents of action instances and identify the corresponding categories with only a single point label for each action instance for training. Due to the sparse frame-level annotations, most existing models are in the localization-by-classification pipeline. However, there exist two major issues in this pipeline: large intra-action variation due to task gap between classification and localization and noisy classification learning caused by unreliable pseudo training samples. In this paper, we propose a novel framework CRRC-Net, which introduces a co-supervised feature learning module and a probabilistic pseudo label mining module, to simultaneously address the above two issues. Specifically, the co-supervised feature learning module is applied to exploit the complementary information in different modalities for learning more compact feature representations. Furthermore, the probabilistic pseudo label mining module utilizes the feature distances from action prototypes to estimate the likelihood of pseudo samples and rectify their corresponding labels for more reliable classification learning. Comprehensive experiments are conducted on different benchmarks and the experimental results show that our method achieves favorable performance with the state-of-the-art.
Keyword:
Compact representation
reliable classification
point-level weakly-supervised action localization
期刊
IF:
13.7
论文数:
1.0W
被引数:
8.4W
机构
引用论文
Adsorption of an Au atom and dimer on a thin θ-Al2O3/NiAl(100) film: dependence on the thickness of the θ-Al2O3 film.png)
RSC Advances
OAAI
RSC AdvancesIF0
OAAI分享收藏
查看更多内容