Skip to content

This repository hosts code and resources for a project using Siamese Neural Networks (SNN) for seismic data analysis in geophysics. The goal is to develop an algorithm that identifies and matches similar seismic features, assisting geoscientists in interpreting subsurface geological structures.

License

Notifications You must be signed in to change notification settings

association-rosia/reflection-connection

Repository files navigation

🪨 Reflection Connection

This project uses Siamese Neural Networks (SNN) to analyze seismic data, aiding geoscientists in identifying subsurface geological structures. SNNs are trained on a limited dataset of seismic features to recognize and match similar patterns within a larger corpus. This approach reduces manual effort and error, empowering researchers to focus on integrating geological insights into comprehensive models of the Earth's subsurface.

This project was made possible by our compute partners 2CRSi and NVIDIA.

🏆 Challenge ranking

The score of the challenge was a custom score. The evaluation metric used for scoring will be weighted accuracy:

image

The final score for all N query images is then calculated as follows:

image

Our solution was the 8th (out of 147 teams) one with a score equal to 0.4001 🎉.

Prediction Leaderboard podium:
🥇 DAISUKE YAMAMOTO - 0.5168
🥈 Igor Ivanov - 0.4936
🥉 Dmitry Ulyanov - 0.4500

🏛️ Proposed solution

Pretraining:

Fine-tuning (iterative training):

#️⃣ Command lines

Launch a training

python src/models/<model path>/train_model.py

View project's runs on WandB.

🔬 References

Oquab, M., Darcet, T., Moutakanni, T., Vo, H., Szafraniec, M., Khalidov, V., ... & Bojanowski, P. (2023). Dinov2: Learning robust visual features without supervision. arXiv preprint arXiv:2304.07193.

He, K., Chen, X., Xie, S., Li, Y., Dollár, P., & Girshick, R. (2022). Masked autoencoders are scalable vision learners. In Proceedings of the IEEE/CVF conference on computer vision and pattern recognition (pp. 16000-16009).

Radford, A., Kim, J. W., Hallacy, C., Ramesh, A., Goh, G., Agarwal, S., ... & Sutskever, I. (2021, July). Learning transferable visual models from natural language supervision. In International conference on machine learning (pp. 8748-8763). PMLR.

📝 Citing

@misc{UrgellReberga:2024,
  Author = {Baptiste Urgell and Louis Reberga},
  Title = {Reflection Connection},
  Year = {2024},
  Publisher = {GitHub},
  Journal = {GitHub repository},
  Howpublished = {\url{https://github.com/association-rosia/reflection-connection}}
}

🛡️ License

Project is distributed under MIT License

👨🏻‍💻 Contributors

Louis REBERGA

Baptiste URGELL

About

This repository hosts code and resources for a project using Siamese Neural Networks (SNN) for seismic data analysis in geophysics. The goal is to develop an algorithm that identifies and matches similar seismic features, assisting geoscientists in interpreting subsurface geological structures.

Topics

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published