Improving improved photorealism

Developers from Intel Labs use the convolutional neural network to improve synthesized images, increase their stability and realism.

Gta v to Cityscapes

The modifications made by our method are geometrically and semantically consistent with the original images.

The modifications are also stable over time:

Greening Scorched Grass and Hills in California in GTA:

Add reflections to the windows and increase the Fresnel effect (for example, on the roof of cars):

Rebuilding roads:

Translation of GTA V into Mapillary vistas

Images from this dataset are recorded around the world using a wide variety of cameras. The images are brighter and with higher resolution.

Remove the distant haze and rebuild the road:

We make the grass more voluminous and green:

• Data and code: github.com/intel-isl/PhotorealismEnhancement
• Article: Enhancing photorealism enhancement