New RD-SCL framework improves few-shot acoustic impedance imaging

Researchers have developed a new framework called RD-SCL for few-shot acoustic impedance imaging, a challenging problem in subsurface analysis due to limited labeled data and unknown seismic wavelets. This method integrates regularized deconvolution with semi-supervised cross-learning, featuring a differentiable deconvolution operator that dynamically estimates the latent wavelet during training. Experiments on the SEAM and Marmousi 2 benchmarks show RD-SCL outperforms existing methods in accuracy and efficiency, utilizing fewer learnable parameters and competitive runtime. AI

IMPACT This research offers a more efficient and physically consistent solution for subsurface analysis, potentially improving seismic imaging accuracy.

RANK_REASON The cluster contains an academic paper detailing a new method for a specific scientific problem. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.CV →

• arXiv
• Marmousi 2
• RD-SCL
• SEAM

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →