{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/velocitygan-data-driven-full-waveform","title":"VelocityGAN: Data-Driven Full-Waveform Inversion Using Conditional Adversarial Networks","arxiv_id":"1809.10262","date":"2018-09-26","proceeding":null,"authors":["Zhongping Zhang","Yue Wu","Zheng Zhou","Youzuo Lin"],"abstract":"Acoustic- and elastic-waveform inversion is an important and widely used\nmethod to reconstruct subsurface velocity image. Waveform inversion is a\ntypical non-linear and ill-posed inverse problem. Existing physics-driven\ncomputational methods for solving waveform inversion suffer from the cycle\nskipping and local minima issues, and not to mention solving waveform inversion\nis computationally expensive. In this paper, we developed a real-time\ndata-driven technique, VelocityGAN, to accurately reconstruct subsurface\nvelocities. Our VelocityGAN is an end-to-end framework which can generate\nhigh-quality velocity images directly from the raw seismic waveform data. A\nseries of numerical experiments are conducted on the synthetic seismic\nreflection data to evaluate the effectiveness and efficiency of VelocityGAN. We\nnot only compare it with existing physics-driven approaches but also choose\nsome deep learning frameworks as our data-driven baselines. The experiment\nresults show that VelocityGAN outperforms the physics-driven waveform inversion\nmethods and achieves the state-of-the-art performance among data-driven\nbaselines.","url_abs":"http://arxiv.org/abs/1809.10262v4","url_pdf":"http://arxiv.org/pdf/1809.10262v4.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"velocitygan-data-driven-full-waveform","repo_url":"https://github.com/lanl/openfwi","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"BSD-3-Clause"}},{"paper_slug":"velocitygan-data-driven-full-waveform","repo_url":"https://github.com/PaddlePaddle/PaddleScience","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"paddle","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1809.10262","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}