Improved training of binary networks for human pose estimation and image recognition

11 Apr 2019  ·  Adrian Bulat, Georgios Tzimiropoulos, Jean Kossaifi, Maja Pantic ·

Big neural networks trained on large datasets have advanced the state-of-the-art for a large variety of challenging problems, improving performance by a large margin. However, under low memory and limited computational power constraints, the accuracy on the same problems drops considerable. In this paper, we propose a series of techniques that significantly improve the accuracy of binarized neural networks (i.e networks where both the features and the weights are binary). We evaluate the proposed improvements on two diverse tasks: fine-grained recognition (human pose estimation) and large-scale image recognition (ImageNet classification). Specifically, we introduce a series of novel methodological changes including: (a) more appropriate activation functions, (b) reverse-order initialization, (c) progressive quantization, and (d) network stacking and show that these additions improve existing state-of-the-art network binarization techniques, significantly. Additionally, for the first time, we also investigate the extent to which network binarization and knowledge distillation can be combined. When tested on the challenging MPII dataset, our method shows a performance improvement of more than 4% in absolute terms. Finally, we further validate our findings by applying the proposed techniques for large-scale object recognition on the Imagenet dataset, on which we report a reduction of error rate by 4%.

PDF Abstract
Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Classification with Binary Neural Network ImageNet ResNet Top-1 Accuracy 53.7 # 6
Classification with Binary Neural Network ImageNet AlexNet Top-1 Accuracy 48.6 # 8
Pose Estimation MPII Human Pose Ours PCKh-0.5 80.9 # 36

Methods