逐行源码阅读中文笔记。C3D的实现很简单,对着模型架构直接实现就ok。
blog解读:https://blog.csdn.net/qq_39388410/article/details/104951012
原paper: Tran, Du, et al. "Learning spatiotemporal features with 3d convolutional networks." Proceedings of the IEEE international conference on computer vision. 2015.
参考code: https://github.com/DavideA/c3d-pytorch
另外还有个即插即玩的code也挺不错:https://github.com/jfzhang95/pytorch-video-recognition
或者直接使用商汤OpenMMLab的开源。