Grand Theft Auto V: Enhancing Photorealism Enhancement
GTA 5 Running on Enhancing Photorealism Enhancement (Machine Learning)
Grand Theft Auto 5 is no stranger to photorealism mods, but an AI research group from Intel has taken it to the next level with their AI machine learning tool. The results are amazing and offer up a hint as to what a Grand Theft Auto 5 Remastered could look like.
Intel ISL research group's Photorealism Enhancement Project works by the AI analyzing each frame of the game and then comparing it to a huge database of photographs, in which it then applies various enhancements to the footage.
The images are enhanced by a convolutional network that leverages intermediate representations produced by conventional rendering pipelines. The network is trained via a novel adversarial objective, which provides strong supervision at multiple perceptual levels. We analyze scene layout distributions in commonly used datasets and find that they differ in important ways. We hypothesize that this is one of the causes of strong artifacts that can be observed in the results of many prior methods. To address this we propose a new strategy for sampling image patches during training. We also introduce multiple architectural improvements in the deep network modules used for photorealism enhancement. We confirm the benefits of our contributions in controlled experiments and report substantial gains in stability and realism in comparison to recent image-to-image translation methods and a variety of other baselines.
Disclaimer! This video has been made by using ReShade 4.9.1 tools
More information:
https://intel-isl.github.io/PhotorealismEnhancement
Other Videos By Vespucci
Other Statistics
Grand Theft Auto V Statistics For Vespucci
Currently, Vespucci has 649,263 views for Grand Theft Auto V across 75 videos. There's close to 10 hours worth of content for Grand Theft Auto V published on his channel, or 21.05% of the total watchable video on Vespucci's YouTube channel.