Yet Another Computer Vision Index To Datasets (YACVID) - Details

Stand: 2019-07-18 000000m 13:05:35 - Overview

ok42
Attribute Current Content New
Name (Institute + Shorttitle)MultiDrone Public DataSet 
Description (include details on usage, files and paper references)The public MultiDrone Dataset has been assembled using both pre-existing audiovisual material and newly filmed UAV shots. A large subset of these data has been annotated for facilitating scientific research, in tasks such as visual detection and tracking of bicycles, football players, human crowds, etc. The following list gives an overview over the available data and respective details of the full dataset:

1)More than 10 hours of UAV footage depicting football, rowing and cycling (DW)

2)More than 115 GBs of professional aerial/UAV shots depicting bicycle races (RAI, DW, AUTH)

3)More than 2.5 GBs of general, professional and semi-professional UAV shots in varied environments (AUTH)

4)More than 32 GBs of aerial/UAV shots depicting human crowds (AUTH)

5)More than 2 GBs of aerial/UAV shots depicting boat races (AUTH, RAI)

6)More than 6 GBs of ground shots depicting a UAV flying (AUTH)

7)More than 2.5 GBs of UAV shots depicting buildings-of-interest (AUTH)

8)Additional UAV datasets for face de-identification, person detection, potential landing site detection (AUTH)

9)More than 25 minutes of simulation videos using a state-of-the-art real-time 3D graphics engine, used to characterise optimal drone parameters for specific scenarios and shot types in terms of viewing experience. (UoB)

10)Skeletal joints of all the detected people (including cyclists and spectators) corresponding to more than 2 hours of footage (IST)

For a detailed overview of the individual data sets, download our dataset description here. The sheet lists all available sets, including a description, the responsible partner as well as the proper licences. Unless otherwise noted, these datasets are stored on the MultiDrone Media Gallery. Appropriate credentials can be requested via our partner RAI (A. Messina). 
URL Link120586240 
Files (#)
References (SKIPPED)
Category (SKIPPED) 
Tags (single words, spaced)drone uav image tracking sport football rowing bicycle human crowd boat race 
Last Changed2019-07-18 
Turing (2.12+3.25=?) :-)