High-Fidelity Compression of Seismic Velocity Models via SIREN Auto-Decoders

2026-03-15

AI Summary

This work proposes an implicit neural representation framework based on a SIREN auto-decoder to address the challenge of efficiently compressing and faithfully reconstructing complex, multi-structured seismic velocity models. The method encodes each 70×70 velocity map into a 256-dimensional latent vector, achieving a 19:1 compression ratio while enabling zero-shot super-resolution reconstruction at arbitrary resolutions—up to 280×280—and geologically plausible interpolation in the latent space. Evaluated on the OpenFWI dataset, the approach achieves an average PSNR of 32.47 dB and SSIM of 0.956, surpassing the limitations of conventional grid-dependent representations and significantly enhancing both compression efficiency and multi-scale applicability.

Technology Category

• Machine Learning: Deep Generative Models & Autoencoders
• Computer Vision: Representation Learning for Vision
• Knowledge Representation and Reasoning: Geometric, Spatial, and Temporal Reasoning

Application Category

• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• Graph Algorithms and Modeling for the Web: Graph embeddings and representation learning for Web-related graphs
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics

Abstract

Implicit Neural Representations (INRs) have emerged as a powerful paradigm for representing continuous signals independently of grid resolution. In this paper, we propose a high-fidelity neural compression framework based on a SIREN (Sinusoidal Representation Networks) auto-decoder to represent multi-structural seismic velocity models from the OpenFWI benchmark. Our method compresses each 70x70 velocity map (4,900 points) into a compact 256-dimensional latent vector, achieving a compression ratio of 19:1. We evaluate the framework on 1,000 samples across five diverse geological families: FlatVel, CurveVel, FlatFault, CurveFault, and Style. Experimental results demonstrate an average PSNR of 32.47 dB and SSIM of 0.956, indicating high-quality reconstruction. Furthermore, we showcase two key advantages of our implicit representation: (1) smooth latent space interpolation that generates plausible intermediate velocity structures, and (2) zero-shot super-resolution capability that reconstructs velocity fields at arbitrary resolutions up to 280x280 without additional training. The results highlight the potential of INR-based auto-decoders for efficient storage, multi-scale analysis, and downstream geophysical applications such as full waveform inversion.

Problem

• seismic velocity models
• high-fidelity compression
• implicit neural representations
• neural compression
• data representation

Innovation

• Implicit Neural Representation
• SIREN Auto-Decoder
• Seismic Velocity Model Compression
• Zero-Shot Super-Resolution
• Latent Space Interpolation