Terrain Generation with Deep Learning
Subscribe:  iCal  |  Google Calendar
7, Mar — 12, Jun
Austin US   9, Mar — 19, Mar
San Francisco US   19, Mar — 24, Mar
San Francisco US   19, Mar — 21, Mar
Anaheim US   23, Mar — 26, Mar
Latest comments

Donald Trump, insulation is a seamless wall with airpockets. Ceilings can be printed using a re-enforcing scaffold for support. Try googling info..

by Polygrinder
8 hours ago

Really awesome work and the tutorial is fantastic. Thanks for sharing.

by Dave
8 hours ago

Absolutely no information about the 4.2 release - was it ever released in September. There is about as much information on trueSKY as there is in any of the so called products that use it. For me this lack of transparency is killing there business and points to fundamental issues with the technology. Google trueSKY in YouTube and you'll hardly get any information at all. For such a ground breaking technology this is very suspicious. Do they not have a marketing team - do they even care? Sounds like a very small company which wishes to remain small and doesn't understand what they can become because with the technology they have they should be targeting a bigger profile, revenue streams and audiance than they have and the lack of foresight here with the Simul management is quite frankly very disapointing. Another 10 years could easily disapear for these guys and they will simply remain a small fish. Very sad.

Terrain Generation with Deep Learning
9 October, 2017

Eric Guérin has recently presented a new model that makes an attempt to build complete terrain maps from a few user sketches. The thing can potentially change the way we deal with landscape generation, so let’s start studying it. 

The model was originally described in a paper called “Interactive Example-Based Terrain Authoring with Conditional Generative Adversarial Networks” by Eric Guérin, Julie Digne, Eric Galin, Adrien Peytavie, Christian Wolf, Bedrich Benes and Benoît Martinez. You can read about the idea below.


Authoring virtual terrains presents a challenge and there is a strong need for authoring tools able to create realistic terrains with simple user-inputs and with high user control. We propose an example-based authoring pipeline that uses a set of terrain synthesizers dedicated to specific tasks. Each terrain synthesizer is a Conditional Generative Adversarial Network trained by using real-world terrains and their sketched counterparts. The training sets are built automatically with a view that the terrain synthesizers learn the generation from features that are easy to sketch. During the authoring process, the artist first creates a rough sketch of the main terrain features, such as rivers, valleys and ridges, and the algorithm automatically synthesizes a terrain corresponding to the sketch using the learned features of the training samples. Moreover, an erosion synthesizer can also generate terrain evolution by erosion at a very low computational cost. Our framework allows for an easy terrain authoring and provides a high level of realism for a minimum sketch cost. We show various examples of terrain synthesis created by experienced as well as inexperienced users who are able to design a vast variety of complex terrains in a very short time.

Get the full text

Source: LinkedIn

Leave a Reply

Be the First to Comment!