用对抗策略选关键样本，让骨骼动作识别少要标签仍高效
Active Learning for GCN-based Action Recognition
- 用对抗思想选最有信息量的样本，兼顾代表性、多样性和不确定性
- 在两个骨架动作识别数据集上，用更少标签达到更好效果
- 适合数据标注成本高、需高效学习的场景
尽管图卷积网络（GCNs）在基于骨骼的动作识别中表现优异，但其性能通常依赖大量标注数据，而实际场景中标签往往稀缺。为此，我们提出一种新型标签高效的GCN模型。首先，设计了一种新颖的获取函数，采用对抗策略筛选出少量但信息丰富的样本用于标注，该过程平衡了代表性、多样性和不确定性。其次，引入双向且稳定的GCN架构，增强环境空间与潜在数据空间之间的映射能力，有助于更好地理解所学样本的分布特性。在两个具有挑战性的骨架动作识别基准数据集上的广泛实验表明，相比先前方法，我们的标签高效GCN取得了显著提升。
原文摘要 · Abstract (English)
Despite the notable success of graph convolutional networks (GCNs) in skeleton-based action recognition, their performance often depends on large volumes of labeled data, which are frequently scarce in practical settings. To address this limitation, we propose a novel label-efficient GCN model. Our work makes two primary contributions. First, we develop a novel acquisition function that employs an adversarial strategy to identify a compact set of informative exemplars for labeling. This selection process balances representativeness, diversity, and uncertainty. Second, we introduce bidirectional and stable GCN architectures. These enhanced networks facilitate a more effective mapping between the ambient and latent data spaces, enabling a better understanding of the learned exemplar distribution. Extensive evaluations on two challenging skeleton-based action recognition benchmarks reveal significant improvements achieved by our label-efficient GCNs compared to prior work.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。