Contrastive Learning for Seismic Horizon Tracking with Domain-Specific Priors
2026-06-15 • Computer Vision and Pattern Recognition
Computer Vision and Pattern RecognitionMachine Learning
AI summaryⓘ
The authors address challenges in tracking 3D seismic horizons without labels. They combine two methods: one that aligns data trace-by-trace but struggles near faults, and another deep learning method that handles faults better but usually needs labeled data. Their approach uses reliable signal-based matches to guide training of a texture-focused neural network, helping it learn to follow horizons even through difficult fault areas. Tests show their method reduces errors compared to other unsupervised approaches and performs similarly to some semi-supervised methods.
3D seismic horizon trackingunsupervised learningself-supervised learningcontrastive learningreflector slopesfaults in seismic dataneural network embeddingstexture-based deep learningmean absolute error (MAE)
Authors
Alexandre Thouvenot, Lionel Boillot, Vincent Gripon
Abstract
Unsupervised 3D seismic horizon tracking faces a key limitation: signal-based propagators provide accurate trace-level alignment but often fail near faults, whereas texture-driven deep models are more robust to discontinuities, typically at the cost of labeled data requirements and reduced trace-level precision. We propose a self-supervised fusion of both paradigms in which signal-derived local horizon correspondences act as domain-specific priors to train a texture-based deep learning model. Specifically, we estimate reliable trace-to-trace flows from reflector slopes and use them to form positive pairs in a contrastive objective, while restricting training to high-confidence neighborhoods, optionally augmented with a fault mask. The objective is not to infer ambiguous correspondences close to discontinuities, but to preserve horizon identity across them. As a result, the network learns voxel-wise embeddings that preserve local signal continuity while enabling horizon propagation beyond discontinuities through similarity search. Experiments on the public F3 dataset and a faulted synthetic dataset achieve lower mean absolute error (MAE) than unsupervised baselines and competitive performance against a semi-supervised method using a single labeled slice.