Efficient Seismic Data Interpolation via Sparse Attention Transformer and Diffusion Model

TL;DR

This paper introduces Diff-spaformer, a novel deep learning framework combining sparse attention transformers and diffusion models to improve seismic data interpolation efficiency and accuracy, especially under complex geological conditions.

Contribution

The paper presents a new diffusion-enhanced sparse attention transformer that integrates transformer and diffusion models via a Seismic Prior Extraction Network for efficient seismic data interpolation.

Findings

Improved interpolation fidelity over existing methods.

Enhanced computational efficiency in seismic data reconstruction.

Effective handling of complex geological data conditions.

Abstract

Seismic data interpolation is a critical pre-processing step for improving seismic imaging quality and remains a focus of academic innovation. To address the computational inefficiencies caused by extensive iterative resampling in current plug-and-play diffusion interpolation methods, we propose the diffusion-enhanced sparse attention transformer (Diff-spaformer), a novel deep learning framework. Our model integrates transformer architectures and diffusion models via a Seismic Prior Extraction Network (SPEN), which serves as a bridge module. Full-layer sparse multi-head attention and feed-forward propagation capture global information distributions, while the diffusion model provides robust prior guidance. To mitigate the computational burden of high-dimensional representations, self-attention is computed along the channel rather than the spatial dimension. We show that using negative squared Euclidean distance to compute sparse affinity matrices better suits seismic data modeling, enabling broader contribution from amplitude feature nodes. An adaptive ReLU function further discards low or irrelevant self-attention values. We conduct training within a single-stage optimization framework, requiring only a few reverse diffusion sampling steps during inference. Extensive experiments demonstrate improved interpolation fidelity and computational efficiency for both random and continuous missing data, offering a new paradigm for high-efficiency seismic data reconstruction under complex geological conditions.