Three-dimensional convolutional neural networks (3DCNNs) have been applied in many tasks, e.g., video and 3D point cloud recognition. However, due to the higher dimension of convolutional kernels, the space complexity of 3DCNNs is generally larger than that of traditional two-dimensional convolutional neural networks (2DCNNs). To miniaturize 3DCNNs for the deployment in confining environments such as embedded devices, neural network compression is a promising approach. In this work, we adopt the tensor train (TT) decomposition, a straightforward and simple in situ training compression method, to shrink the 3DCNN models. Through proposing tensorizing 3D convolutional kernels in TT format, we investigate how to select appropriate TT ranks for achieving higher compression ratio. We have also discussed the redundancy of 3D convolutional kernels for compression, core significance and future directions of this work, as well as the theoretical computation complexity versus practical executing time of convolution in TT. In the light of multiple contrast experiments based on VIVA challenge, UCF11, UCF101, and ModelNet40 datasets, we conclude that TT decomposition can compress 3DCNNs by around one hundred times without significant accuracy loss, which will enable its applications in extensive real world scenarios.

三维卷积神经网络（3DCNN）已应用于许多任务，例如视频和3D点云识别。但是，由于卷积核的维数较大，因此3DCNN的空间复杂度通常比传统的二维卷积神经网络（2DCNN）大。为了使3DCNN小型化以在受限环境（例如嵌入式设备）中进行部署，神经网络压缩是一种有前途的方法。在这项工作中，我们采用张量链（TT）分解，一种简单直接的原位训练压缩方法，以缩小3DCNN模型。通过提出张量化TT格式的3D卷积核，我们研究了如何选择合适的TT等级以获得更高的压缩率。我们还讨论了3D卷积内核在压缩方面的冗余性，核心意义和这项工作的未来方向，以及理论计算复杂度与TT中卷积的实际执行时间。根据基于VIVA挑战，UCF11，UCF101和ModelNet40数据集的多次对比实验，我们得出的结论是，TT分解可以将3DCNN压缩大约100倍而没有明显的精度损失，这将使其能够在广泛的实际场景中使用。