Cloud gaming dataset - PC gameplay video traces annotated with video metrics

Skup podataka

• Autor: Ivan Slivar; University of Zagreb, Faculty of Electrical Engineering and Computing
• Znanstveno / umjetničko područje, polje i grana: TEHNIČKE ZNANOSTI; Računarstvo; Arhitektura računalnih sustava

Sažetak (engleski)

Large and openly available dataset of 225 different gameplay videos recorded across 25 different games and annotated with objective video metrics.

• Metodologija (engleski): Video game traces were collected in a laboratory environment. Valve’s Steam In-Home streaming platform was used as the cloud gaming environment. The Steam In-Home Streaming client in this case was installed on a HP Probook 4530s laptop, while Steam In-Home Streaming server was installed on a Windows PC desktop. The PC server and laptop client were connected via a wireless access point (both the PC server and the laptop client have a wired connection). The FRAPS application was used to record gameplay sessions. All video traces were recorded at a video encoding frame rate of 30 fps and 10 Mbps video bitrate. Tested games were played in HD-ready resolution (720p) with default graphics settings. For each of the tested games, between 5-10 gameplay video traces were recorded that lasted exactly 30 seconds each in order to obtain a large enough sample of gameplay for each game. Alongside gameplay recording, the intensity of user interaction was measured, thus collecting mouse and keyboard input during gameplay by using the Mousotron application . With respect to tested games, gaming sessions of 25 different video games were recorded. As a result, 225 different video traces were gathered. For each of the recorded video traces, the following temporal and spatial characteristics were calculated: Spatial perceptual information (SI), Temporal perceptual information (TI), Percentage of Forward/backward or Intra-coded Macroblocks (PFIM) and Intra-coded Block Size (IBS).
• Ključne riječi (engleski): Cloud gaming; Video gameplay; Steam In-Home Streaming; SI; TI; PFIM; IBS
• Jezik: engleski
• URN:NBN: urn:nbn:hr:168:250898
• Datum objave: 2021-04-19
• Verzija: 1.0
• Geolokacija: Zagreb, Hrvatska
• Projekt: Šifra: UIP-2014-09-5605; Naziv (hrvatski): Kooperativno upravljanje iskustvenom kvalitetom upokretnim mrežama za interaktivne višemedijske aplikacije u računalom oblaku; Naziv (engleski): Cooperative Quality of Experience Management for Interactive Cloud-Based Multimedia Applications in Mobile Networks; Kratica: Q-MANIC; Voditelj: Lea Skorin-Kapov; Pravna nadležnost: Hrvatska; Financijer: HRZZ; Linija financiranja: UIP
• Vanjski povezani objekt: Sadrži dio: DOI https://doi.org/10.1145/3132041
• Izdavač: Fakultet elektrotehnike i računarstva; Faculty of Electrical Engineering and Computing
• Prava pristupa: Otvoreni pristup
• Datum i vrijeme pohrane: 2021-04-19 13:31:34