Training a Binary Weight Object Detector by Knowledge Transfer for Autonomous Driving
arXiv: Computer Vision and Pattern Recognition(2019)
摘要
Autonomous driving has harsh requirements of small model size and energy efficiency, in order to enable the embedded system to achieve real-time on-board object detection. Recent deep convolutional neural network based object detectors have achieved state-of-the-art accuracy. However, such models are trained with numerous parameters and their high computational costs and large storage prohibit the deployment to memory and computation resource limited systems. Low-precision neural networks are popular techniques for reducing the computation requirements and memory footprint. Among them, binary weight neural networks (BWNs) are the extreme case which quantizes the float-point into just 1 bit. BWNs are difficult to train and suffer from accuracy deprecation due to the extreme low-bit representation. To address this problem, we propose a knowledge transfer (KT) method to aid the training of BWN using a full-precision teacher network. We built DarkNet- and MobileNet-based binary weight YOLOv2 detectors and conduct experiments on KITTI benchmark for car, pedestrian and cyclist detection. The experimental results show that the proposed method maintains high detection accuracy while reducing the model size of DarkNet-YOLO from 257 MB to 8.8 MB and MobileNet-YOLO from 193 MB to 7.9 MB.
更多查看译文
关键词
binary weight object detector,autonomous driving,energy efficiency,on-board object detection,object detectors,low-precision neural networks,computation requirements,memory footprint,binary weight neural networks,BWNs,knowledge transfer method,full-precision teacher network,MobileNet-based binary weight YOLOv2 detectors,pedestrian,cyclist detection,KITTI benchmark,MobileNet-YOLO,DarkNet-YOLO,deep convolutional neural network,word length 1.0 bit,memory size 8.8 MByte to 257.0 MByte,memory size 7.9 MByte to 193.0 MByte
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络