Quantized Densely Connected U-Nets for Efficient Landmark Localization
arXiv:1808.02194
Abstract
In this paper, we propose quantized densely connected U-Nets for efficient visual landmark localization. The idea is that features of the same semantic meanings are globally reused across the stacked U-Nets. This dense connectivity largely improves the information flow, yielding improved localization accuracy. However, a vanilla dense design would suffer from critical efficiency issue in both training and testing. To solve this problem, we first propose order-K dense connectivity to trim off long-distance shortcuts; then, we use a memory-efficient implementation to significantly boost the training efficiency and investigate an iterative refinement that may slice the model size in half. Finally, to reduce the memory consumption and high precision operations both in training and testing, we further quantize weights, inputs, and gradients of our localization network to low bit-width numbers. We validate our approach in two tasks: human pose estimation and face alignment. The results show that our approach achieves state-of-the-art localization accuracy, but using ~70% fewer parameters, ~98% less model size and saving ~75% training memory compared with other benchmark localizers. The code is available at https://github.com/zhiqiangdon/CU-Net.
arXiv commentECCV2018
References in corpus (14)
- Communication-Efficient Learning of Deep Networks from Decentralized Data
- DeepPose: Human Pose Estimation via Deep Neural Networks
- Binarized Neural Networks: Training Deep Neural Networks with Weights and Activations Constrained to +1 or -1
- DoReFa-Net: Training Low Bitwidth Convolutional Neural Networks with Low Bitwidth Gradients
- Ternary Weight Networks
- Human pose estimation via Convolutional Part Heatmap Regression
- Stacked Hourglass Networks for Human Pose Estimation
- Training and Inference with Integers in Deep Neural Networks
- Convolutional Pose Machines
- Learning Feature Pyramids for Human Pose Estimation
- Jointly Optimize Data Augmentation and Network Training: Adversarial Data Augmentation in Human Pose Estimation
- DeepRebirth: Accelerating Deep Neural Network Execution on Mobile Devices
- Human Pose Estimation using Deep Consensus Voting
- Deep Regression for Face Alignment