Machine learning has come a long way and can offer many benefits when it comes to creating art for games, according to So Real Digital Twins head of game development Kelly Vero.
Speaking at PGC Digital #2 on 'Working Smart - Machine Learning in Art Asset Creation for the Games Industry' as part of the Developer Toolkit track, Vero outlined a number of practical use cases from her experience in the industry.
These include asset storage and creation vertex/object examination/analysis, pipeline performance and QA. Any subset of artificial can make developer's lives "a while lot simpler" says Vero, though she notes it's still early on and the technology has a long way to go.
"We're creating a solution for a problem for something everyone has or will have," stated Vero.
Image segmentation tools, in particular, were spotlighted for their positive effects on art in games. A process of examine, simplify, speed, learn and predict will help cut down creation times and in theory speed the output process.
Vero will once again be speaking about the topic later today on the PGC Digital track: 'Machine Takeover - Improving Your Game With Machine Learning & AI'.
PG Connects Digital #2 is the best of our Pocket Gamer Connects conference in an online form, with an entire week of talks, meetings, and pitch events taking place from June 8th to the 12th. You can read up on all the tracks taking place through the week here.