Dataset for Detecting Motorcyclists in Pedestrian Areas

The file PedestrianDataSet.zip contains 2 directories and 1 file, the directory named images contains all the images captured; the directory named “etiquetas” contains all annotations done based on four classes ordered as [MMNAP, MMAP, MNAP, PCP], the name's correspondence with the label is [motorcycle with motorcyclist not in pedestrian area, motorcycle with motorcyclist in pedestrian area, motorcycle without motorcyclist in pedestrian area, pedestrian in crosswalk]; the file Distribution.png is an image that explain how the DataSet is distributed.

The dataset contains a 6324 images in .jpg format for a total of 643.9MB, the resolution of each image is 1280x720 (HD), and the weight per image is under 120 KB, the images were obtained from a closed circuit television CCTV free public access located in the city of Medellín Colombia consisting of 80 cameras located on the road, these images were captured mainly during daylight hours. The format of the name of each image is "image (camera number)_(date format"AAAAMMDD")_( time format "hhmmss"), where (AAAA) is the year, (MM) is the month, (DD) is the day, (HH) is the 24-hour hour, (MM) is the minute and (SS) is the second.