Journal article Open Access

DVS Benchmark Datasets for Object Tracking, Action Recognition, and Object Recognition

Hu, Yuhuang; Liu, Hongjie; Pfeiffer, Michael; Delbruck, Tobi

This data report summarizes a new benchmark dataset in which we converted established visual video benchmarks for object tracking, action recognition and object recognition into spiking neuromorphic datasets, recorded with the DVS output (Lichtsteiner et al., 2008) of a DAVIS camera (Berner et al., 2013; Brandli et al., 2014). This report presents our approach for sensor calibration and capture of frame-based videos into neuromorphic vision datasets with minimal human intervention. We converted four widely used dynamic datasets: the VOT Challenge 2015 Dataset (Kristan et al., 2016), TrackingDataset3, the UCF-50 Action Recognition Dataset (Reddy and Shah, 2012), and the Caltech-256 Object Category Dataset (Griffin et al., 2006). We conclude with statistics and summaries of the datasets.

 

This repository provides the HDF archives of the dataset.

This research is supported by the European Commission project VISUALISE (FP7-ICT-600954), SeeBetter (FP7-ICT-270324), and the Samsung Advanced Institute of Technology.
Files (26.2 GB)
Name Size
calibration.zip
md5:9f218ab5ca4c9e6bf8234e7564c1c99b
943 Bytes Download
groundtruth-for-vot-and-tracking-dataset-20160610.zip
md5:34d04bb212e5f924568e01df7fd6dd37
1.1 MB Download
INI_Caltech256_10fps_20160424.hdf5.tar.gz
md5:4c48367695d7dd278ada83be62cabd7a
8.1 GB Download
INI_TrackingDataset_30fps_20160610.hdf5.tar.gz
md5:3d47463f7bcee9282550217f6082b757
530.8 MB Download
INI_UCF50_30fps_20160424.hdf5.tar.gz
md5:9d64298e301d77a8c275b6af0777e5dd
17.2 GB Download
INI_VOT_30fps_20160610.hdf5.tar.gz
md5:904e1005cd7572869c49b7a71922ff84
358.8 MB Download
md5_info.txt
md5:7bfe485161d29e875dcb7dd7c5c198cd
896 Bytes Download
37
20
views
downloads
Views 37
Downloads 20
Data volume 69.6 GB
Unique views 28
Unique downloads 14

Share

Cite as