Results - Hackathon 2017
Training Synthetics
The model was trained on 200x200 greyscale velocity models and synthetic seismic models created by 1D convolution with a Ricker wavelet of frequency 10Hz. These are examples from the training dataset.
Input Model
Output Seismic
Reference Seismic
Input Seismic
Output Model
Reference Model
Marmousi Forward (Model 2 Seismic)
The “industry standard” Marmousi model was used as a test case, to validate our models ability to generalise for geometries that it has not seen before. (Image size 2000 px) Notice that very large and arbitrary shaped domains can be mapped. Processing time 5 seconds.
Input Model
Generated Seismic
Marmousi Inversion (Seismic 2 Model)
The “industry standard” Marmousi model was used as a test case, to validate our models ability to generalise for geometries that it has not seen before. (Image size 2000 px) Notice that very large and arbitrary shaped domains can be mapped. Processing time 5 seconds.
Input Seismic
Inverted Model
Porcupine Bay Inversion (Seismic 2 Model)
This is real seismic data plucked for the web, we do not know how this was imaged nor does the system!
Real Seismic
Inverted Model
Results - EAGE 2018
The follow results are generated from a system re-trained using the same "industry standard" on the Marmousi model dataset. (Huge shout to Wouter Kimman who did the retraining work)
Training Examples from the Marmousi model
Testing on f3 Netherlands Dataset
Input Seismic
Inverted Model
