{"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/depth-estimation-from-4d-light-field-videos","title":"Depth estimation from 4D light field videos","arxiv_id":"2012.03021","date":"2020-12-05","proceeding":null,"authors":["Takahiro Kinoshita","Satoshi Ono"],"abstract":"Depth (disparity) estimation from 4D Light Field (LF) images has been a research topic for the last couple of years. Most studies have focused on depth estimation from static 4D LF images while not considering temporal information, i.e., LF videos. This paper proposes an end-to-end neural network architecture for depth estimation from 4D LF videos. This study also constructs a medium-scale synthetic 4D LF video dataset that can be used for training deep learning-based methods. Experimental results using synthetic and real-world 4D LF videos show that temporal information contributes to the improvement of depth estimation accuracy in noisy regions. Dataset and code is available at: https://mediaeng-lfv.github.io/LFV_Disparity_Estimation","url_abs":"https://arxiv.org/abs/2012.03021v2","url_pdf":"https://arxiv.org/pdf/2012.03021v2.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":"depth-estimation-from-4d-light-field-videos","repo_url":"https://github.com/mediaeng-lfv/LFV_Disparity_Estimation","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"depth-estimation","task_name":"Depth Estimation"},{"task_slug":"disparity-estimation","task_name":"Disparity Estimation"}],"methods":[{"method_slug":"3d-convolution","method_name":"3D Convolution"},{"method_slug":"convlstm","method_name":"ConvLSTM"},{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[{"slug":"sintel-4d-lfv","name":"Sintel 4D LFV","full_name":"Sintel 4D Light Field Video Dataset"}],"methods_introduced":[],"results":[{"leaderboard":"/sota/disparity-estimation-on-sintel-4d-lfv","task":"Disparity Estimation","dataset":"Sintel 4D LFV - ambushfight5","model":"Two-stream CNN+CLSTM","rank_in_archive_order":1,"of":1,"metrics":{"BadPix(0.01)":"62.0493","BadPix(0.03)":"22.8762","BadPix(0.07)":"8.3404","MSE*100":"21.67"},"uses_additional_data":false},{"leaderboard":"/sota/disparity-estimation-on-sintel-4d-lfv-bamboo3","task":"Disparity Estimation","dataset":"Sintel 4D LFV - bamboo3","model":"Two-stream CNN+CLSTM","rank_in_archive_order":1,"of":1,"metrics":{"BadPix(0.01)":"53.2985","BadPix(0.03)":"21.8162","BadPix(0.07)":"8.9475","MSE*100":"21.59"},"uses_additional_data":false},{"leaderboard":"/sota/disparity-estimation-on-sintel-4d-lfv-shaman2","task":"Disparity Estimation","dataset":"Sintel 4D LFV - shaman2","model":"Two-stream CNN+CLSTM","rank_in_archive_order":1,"of":1,"metrics":{"BadPix(0.01)":"74.7733","BadPix(0.03)":"50.6706","BadPix(0.07)":"32.7585","MSE*100":"2.4421"},"uses_additional_data":false},{"leaderboard":"/sota/disparity-estimation-on-sintel-4d-lfv-1","task":"Disparity Estimation","dataset":"Sintel 4D LFV - thebigfight2","model":"Two-stream CNN+CLSTM","rank_in_archive_order":1,"of":1,"metrics":{"BadPix(0.01)":"17.7493","BadPix(0.03)":"3.6084","BadPix(0.05)":"1.0688","MSE*100":"3.67"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2012.03021","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}