2.5D Transformer: An Efficient 3D Seismic Interpolation Method without Full 3D Training

TL;DR

This paper introduces a 2.5D Transformer network for 3D seismic data interpolation that leverages cross-dimensional transfer learning, enabling efficient 3D modeling without full 3D training.

Contribution

It proposes a novel T-2.5D architecture combining 2D Transformer encoders with 3D seismic adapters, utilizing a two-stage transfer learning strategy for efficient 3D seismic interpolation.

Findings

Achieves comparable performance to full 3D Transformers

Reduces computational cost significantly

Effective on multiple seismic datasets

Abstract

Transformer has emerged as a powerful deep-learning technique for two-dimensional (2D) seismic data interpolation, owing to its global modeling ability. However, its core operation introduces heavy computational burden due to the quadratic complexity, hindering its further application to higher-dimensional data. To achieve Transformer-based three-dimensional (3D) seismic interpolation, we propose a 2.5-dimensional Transformer network (T-2.5D) that adopts a cross-dimensional transfer learning (TL) strategy, so as to adapt the 2D Transformer encoders to 3D seismic data. The proposed T-2.5D is mainly composed of 2D Transformer encoders and 3D seismic dimension adapters (SDAs). Each 3D SDA is placed before a Transformer encoder to learn spatial correlation information across seismic lines. The proposed cross-dimensional TL strategy comprises two stages: 2D pre-training and 3D fine-tuning. In the first stage, we optimize the 2D Transformer encoders using a large amount of 2D data patches. In the second stage, we freeze the 2D Transformer encoders and fine-tune the 3D SDAs using limited 3D data volumes. Extensive experiments on multiple datasets are conducted to assess the effectiveness and efficiency of T-2.5D. Experimental results demonstrate that the proposed method achieves comparable performance to that of full 3D Transformer at a significantly low cost.