Compressing 3DCNNs Based on Tensor Train Decomposition

8 Dec 2019  ·  Dingheng Wang, Guangshe Zhao, Guoqi Li, Lei Deng, Yang Wu ·

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, and UCF101 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.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Quantization CIFAR-10 3DCNN_VIVA_3 MAP 160327.04 # 1
Quantization Knowledge-based: 3DCNN_VIVA_5 All 84809664 # 1
Hand Gesture Recognition SHREC 2017 track on 3D Hand Gesture Recognition 3DCNN_VIVA_4 14 gestures accuracy 73121216 # 1
Hand-Gesture Recognition VIVA Hand Gestures Dataset Two 3DCNNs: LRN + HRN [11] Accuracy 77.5 # 1
Hand-Gesture Recognition VIVA Hand Gestures Dataset 3DCNN_VIVA_2 Accuracy-CN -13585591 # 2
Hand-Gesture Recognition VIVA Hand Gestures Dataset 3DCNN_VIVA_1 Accuracy-CN 2303240 # 1
Hand-Gesture Recognition VIVA Hand Gestures Dataset Accuracy 6.86 # 2