SqueezeMap: Fast Pedestrian Detection on a Low-Power Automotive Processor Using Efficient Convolutional Neural Networks.
Rytis VerbickasRobert LaganièreDaniel LarocheChangyun ZhuXiaoyin XuAli OrsPublished in: CVPR Workshops (2017)
Keyphrases
- low power
- pedestrian detection
- high speed
- single chip
- low cost
- power consumption
- human body
- convolutional neural networks
- gate array
- object detection
- human detection
- digital signal processing
- low power consumption
- vlsi circuits
- benchmark datasets
- d objects
- similarity measure
- detection rate
- pairwise
- logic circuits
- power reduction
- multiscale