行为识别论文笔记|I3D S3D R(2+1)D P3D CSN

最新推荐文章于 2023-12-20 15:49:28 发布

Alex丶Chen

最新推荐文章于 2023-12-20 15:49:28 发布

阅读量4.2k

点赞数

分类专栏：视频理解行为识别

本文链接：https://blog.csdn.net/njuptalex/article/details/110366243

版权

这篇博客详细记录了行为识别领域的六种重要模型：I3D、T3D、S3D、R(2+1)D、P3D和CSN的贡献。I3D通过扩展2D卷积到3D以捕捉时空信息；T3D利用2D预训练权重进行知识迁移；S3D提出时空分解卷积；R(2+1)D通过空间时间分解实现损失更快下降；P3D同样采用分解策略，且在效率与准确性间取得平衡；CSN则引入通道分离卷积，降低计算复杂度并保持性能。

摘要生成于 C知道，由 DeepSeek-R1 满血版支持，前往体验 >

行为识别论文笔记-I3D T3D S3D R(2+1)D P3D CSN

I3D

Carreira, Joao, and Andrew Zisserman. “Quo vadis, action recognition? a new model and the kinetics dataset.” proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2017.

T3D

Diba, Ali, et al. “Temporal 3d convnets: New architecture and transfer learning for video classification.” arXiv preprint arXiv:1711.08200 (2017).

S3D

Xie, Saining, et al. “Rethinking spatiotemporal feature learning: Speed-accuracy trade-offs in video classification.” Proceedings of the European Conference on Computer Vision (ECCV). 2018.

R(2+1)D

Tran, Du, et al. “A closer look at spatiotemporal convolutions for action recognition.” Proceedings of the IEEE conference on Computer Vision and Pattern Recognition. 2018.

P3D

Qiu, Zhaofan, Ting Yao, and Tao Mei. “Learn