Pedestrian detection based on faster R-CNN in nighttime by fusing deep convolutional features of successive images.
Jong Hyun KimGanbayar BatchuluunKang Ryoung ParkPublished in: Expert Syst. Appl. (2018)
Keyphrases
- pedestrian detection
- image features
- human body
- detection rate
- test images
- benchmark datasets
- histograms of oriented gradients
- object detection
- input image
- three dimensional
- gabor filters
- image database
- image registration
- feature vectors
- image retrieval
- object recognition
- detection algorithm
- human detection
- feature extraction
- lighting conditions
- keypoints
- color histogram
- image classification
- spatial information
- false positives
- computer vision
- feature set
- multi class
- viewpoint
- support vector
- multiscale
- unsupervised feature learning