| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Predictive Vision Datasets and supporting materials
PVM tracking datasetThis document contains necessary data to reproduce results of the Predictive Vision Paper (http://arxiv.org/abs/1607.06854). The code can be found here. The data necessary to run the code can be downloaded from here in a single large (4GB) zip file. Download the file and unzip it in your home directory. It will create the folder PVM_data where all the data files, simulation files etc. will be stored. The package contains labeled set of movies which are packed as python pickles to use directly with the software. In addition you can download and view those same movies individually as zip files containing individual frames and labels provided in a text file (format widely accepted in the vision tracking literature). There are three sets of movies: green ball, stop sign and face. Each set contains ~40 movies with the respective object in view. The object will typically disappear for some portion of the movie. The data is released under Creative Commons License BY-NC-SA 3.0
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
