datasetposted on 19.09.2020, 04:09 by DEBADITYA ACHARYA
This is the dataset to be used for the parking tutorial using CNN+SVM. The FinalCodeSVM.zip contains sampled PKLot dataset containing 3000 images (empty and occupied), and Barry Street dataset containing 100 images. This folder also contains GroundTruth bounding boxes of Barry street dataset, A ResNet50 feature extractor (for independent operation in MATLAB), the features extracted by ResNet50 for the test images, and a training matrix containing all the features of training images by ResNet50. Please consult the publication and supplementary material for further details
Acharya, D., Yan, W. and Khoshelham, K., 2018, April. Real-time image-based parking occupancy detection using deep learning. In Research@ Locate (pp. 33-40).
Acharya, D. and Khoshelham, K., Real-time image-based parking occupancy detection using deep learning-A CPU friendly MATLAB tutorial. (Supplementary material)