{"418041":{"#nid":"418041","#data":{"type":"news","title":"Georgia Tech Researchers Train Computer to Create Games by Watching YouTube","body":[{"value":"\u003Cp\u003EGeorgia Institute of Technology researchers have developed a computing system that views gameplay video from streaming services like YouTube or Twitch, analyzes the footage and then is able to create original new sections of a game.\u003C\/p\u003E\u003Cp\u003EThe team tested their discovery, the first of its kind, with the original Super Mario Brothers, a well-known two-dimensional platformer game that will allow the new automatic-level designer to replicate results across similar games.\u003C\/p\u003E\u003Cp\u003EThe system focuses on the gaming terrain (not the playable character) and the positioning between elements on-screen \u2013 be it pipes, blocks, coins or Goombas \u2013 and it determines the required relationship or level design rule. For example, pipes in the Mario games tend to stick out of the ground, so the system learns this and prevents any pipes from being flush with grassy surfaces. It also prevents \u201cbreaks\u201d by using spatial analysis \u2013 e.g. no impossibly long jumps for the hero.\u003C\/p\u003E\u003Cp\u003E\u201cAn initial evaluation of our approach indicates an ability to produce level sections that are both playable and close to the original without hand coding any design criteria,\u201d says Matthew Guzdial, lead author and Ph.D. student in Computer Science at Georgia Tech.\u003C\/p\u003E\u003Cp\u003EKey to the process is watching the players in action to see where they actually spend most of their time in the game. After recording on-screen locations of sprites, Georgia Tech\u2019s algorithms determine what are high-interaction areas \u2013 those spots where players spend more time to collect bonus items or master a challenge. The automatic-level designer specifically targets these areas to gain design information. The system is then able to build a new level section, element by element.\u003C\/p\u003E\u003Cp\u003E\u201cOur system creates a model or template, and it\u2019s able to produce level sections that have never been seen before, do not appear random and can be traversed by the player,\u201d says Mark Riedl, the study\u0027s primary investigator and associate professor of Interactive Computing. \u201cOne could say that the system \u2018studies\u2019 the design of Mario levels until it is able to create new playable areas.\u201d\u003C\/p\u003E\u003Cp\u003EThe Georgia Tech system output 151 distinct level sections from 17 samples in the original game, controlling for overall playability and style variables. Output increased to 334 level sections as the system lessened the constraints. The new levels can be played easily by porting them into a game engine.\u003C\/p\u003E\u003Cp\u003ERiedl says this is the first time he is aware of a gameplay video being used to design levels for a Mario game. By applying the technique across a number of different platformer games, a system can theoretically learn genre knowledge, which can be beneficial for procedurally creating games of a given genre. The technique may also extend to other game genres beyond platformers. The researchers next plan to develop full-scale levels and evaluate how gamers interact in those levels compared to the original gameplay videos.\u003C\/p\u003E\u003Cp\u003EThe research, \u201cToward Game Level Generation from Gameplay Videos,\u201d is featured June 22-25 at the Foundations of Digital Games Conference in Pacific Grove, Calif.\u003C\/p\u003E","summary":null,"format":"limited_html"}],"field_subtitle":"","field_summary":[{"value":"\u003Cp\u003EGeorgia Institute of Technology researchers have developed a computing system that views gameplay video from streaming services like YouTube or Twitch, analyzes the footage and then is able to create original new sections of a game.\u003C\/p\u003E\u003Cp\u003EThe team tested their discovery, the first of its kind, with the original Super Mario Brothers, a well-known two-dimensional platformer game that will allow the new automatic-level designer to replicate results across similar games.\u003C\/p\u003E","format":"limited_html"}],"field_summary_sentence":[{"value":"Georgia Institute of Technology researchers have developed a computing system that views gameplay video from streaming services like YouTube or Twitch, analyzes the footage and then is able to create original new sections of a game."}],"uid":"27592","created_gmt":"2015-06-24 11:41:13","changed_gmt":"2016-10-08 03:18:45","author":"Joshua Preston","boilerplate_text":"","field_publication":"","field_article_url":"","dateline":{"date":"2015-06-24T00:00:00-04:00","iso_date":"2015-06-24T00:00:00-04:00","tz":"America\/New_York"},"extras":[],"hg_media":{"418061":{"id":"418061","type":"image","title":"Automatic game level generator","body":null,"created":"1449254269","gmt_created":"2015-12-04 18:37:49","changed":"1475895155","gmt_changed":"2016-10-08 02:52:35","alt":"Automatic game level generator","file":{"fid":"202584","name":"generated_levels_square_set.png","image_path":"\/sites\/default\/files\/images\/generated_levels_square_set_0.png","image_full_path":"http:\/\/www.tlwarc.hg.gatech.edu\/\/sites\/default\/files\/images\/generated_levels_square_set_0.png","mime":"image\/png","size":354991,"path_740":"http:\/\/www.tlwarc.hg.gatech.edu\/sites\/default\/files\/styles\/740xx_scale\/public\/images\/generated_levels_square_set_0.png?itok=xPpYVW6K"}},"418101":{"id":"418101","type":"image","title":"Automatic Game Level Generator","body":null,"created":"1449254269","gmt_created":"2015-12-04 18:37:49","changed":"1475895155","gmt_changed":"2016-10-08 02:52:35","alt":"Automatic Game Level Generator","file":{"fid":"202585","name":"overworldgif.gif","image_path":"\/sites\/default\/files\/images\/overworldgif_0.gif","image_full_path":"http:\/\/www.tlwarc.hg.gatech.edu\/\/sites\/default\/files\/images\/overworldgif_0.gif","mime":"image\/gif","size":6856320,"path_740":"http:\/\/www.tlwarc.hg.gatech.edu\/sites\/default\/files\/styles\/740xx_scale\/public\/images\/overworldgif_0.gif?itok=76RjkuYt"}},"50384":{"id":"50384","type":"image","title":"Mark Riedl","body":null,"created":"1449175392","gmt_created":"2015-12-03 20:43:12","changed":"1475894458","gmt_changed":"2016-10-08 02:40:58","alt":"Mark Riedl","file":{"fid":"128682","name":"mark-riedl.jpg","image_path":"\/sites\/default\/files\/images\/mark-riedl_1.jpg","image_full_path":"http:\/\/www.tlwarc.hg.gatech.edu\/\/sites\/default\/files\/images\/mark-riedl_1.jpg","mime":"image\/jpeg","size":12265,"path_740":"http:\/\/www.tlwarc.hg.gatech.edu\/sites\/default\/files\/styles\/740xx_scale\/public\/images\/mark-riedl_1.jpg?itok=OnWpZU5s"}}},"media_ids":["418061","418101","50384"],"groups":[{"id":"1299","name":"GVU Center"}],"categories":[],"keywords":[],"core_research_areas":[{"id":"39501","name":"People and Technology"}],"news_room_topics":[{"id":"71881","name":"Science and Technology"},{"id":"71901","name":"Society and Culture"}],"event_categories":[],"invited_audience":[],"affiliations":[],"classification":[],"areas_of_expertise":[],"news_and_recent_appearances":[],"phone":[],"contact":[{"value":"\u003Cp\u003E\u003Ca href=\u0022mailto:jpreston@cc.gatech.edu\u0022\u003EJoshua Preston\u003C\/a\u003E \u003Cbr \/\u003EGVU Center, College of Computing \u003Cbr \/\u003E678.231.0787\u003C\/p\u003E","format":"limited_html"}],"email":["jpreston@cc.gatech.edu"],"slides":[],"orientation":[],"userdata":""}}}