arXivDaily arXiv每日学术速递 周一至周五更新
北美退伍军人节(Veterans Day),12 日无更新
arXiv 2407.01397cs.CVcs.AI
收藏

掩码与压缩:持续学习中的高效骨架动作识别

Mask and Compress: Efficient Skeleton-based Action Recognition in Continual Learning

AImageLab - University of Modena and Reggio Emilia(AImageLab - 摩德纳与雷焦艾米利亚大学);Leonardo S.p.A.(莱昂纳多公司)

Matteo Mosconi, Andriy Sorokin, Aniello Panariello, Angelo Porrello, Jacopo Bonato, Marco Cotogni, Luigi Sabetta, Simone Calderara, Rita Cucchiara

AI总结:

针对持续学习场景下的骨架动作识别,提出CHARON框架,通过均匀采样、插值和掩码训练在保持性能的同时降低计算开销,并在Split NTU-60和Split NTU-120上取得新基准。

摘要:

利用骨架数据可以使深度学习模型高效且有效地执行动作识别。在此,我们认为在持续学习的背景下探索这一问题至关重要。尽管大量研究从传统的离线视角关注基于骨架的动作识别,但只有少数研究涉足在线方法。为此,我们提出了CHARON(Continual Human Action Recognition On skeletoNs,基于骨架的持续人类动作识别),它在高效框架内运行的同时保持稳定的性能。通过均匀采样、插值以及基于掩码的内存高效训练阶段等技术,我们在最小化计算开销的同时实现了更高的识别精度。我们在Split NTU-60和所提出的Split NTU-120数据集上的实验表明,CHARON在该领域树立了新的基准。代码可在https://github.com/Sperimental3/CHARON获取。

英文摘要:

The use of skeletal data allows deep learning models to perform action recognition efficiently and effectively. Herein, we believe that exploring this problem within the context of Continual Learning is crucial. While numerous studies focus on skeleton-based action recognition from a traditional offline perspective, only a handful venture into online approaches. In this respect, we introduce CHARON (Continual Human Action Recognition On skeletoNs), which maintains consistent performance while operating within an efficient framework. Through techniques like uniform sampling, interpolation, and a memory-efficient training stage based on masking, we achieve improved recognition accuracy while minimizing computational overhead. Our experiments on Split NTU-60 and the proposed Split NTU-120 datasets demonstrate that CHARON sets a new benchmark in this domain. The code is available at https://github.com/Sperimental3/CHARON.

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。

↑