{"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/stochastic-adversarial-video-prediction","title":"Stochastic Adversarial Video Prediction","arxiv_id":"1804.01523","date":"2018-04-04","proceeding":"ICLR 2019 5","authors":["Alex X. Lee","Richard Zhang","Frederik Ebert","Pieter Abbeel","Chelsea Finn","Sergey Levine"],"abstract":"Being able to predict what may happen in the future requires an in-depth\nunderstanding of the physical and causal rules that govern the world. A model\nthat is able to do so has a number of appealing applications, from robotic\nplanning to representation learning. However, learning to predict raw future\nobservations, such as frames in a video, is exceedingly challenging -- the\nambiguous nature of the problem can cause a naively designed model to average\ntogether possible futures into a single, blurry prediction. Recently, this has\nbeen addressed by two distinct approaches: (a) latent variational variable\nmodels that explicitly model underlying stochasticity and (b)\nadversarially-trained models that aim to produce naturalistic images. However,\na standard latent variable model can struggle to produce realistic results, and\na standard adversarially-trained model underutilizes latent variables and fails\nto produce diverse predictions. We show that these distinct methods are in fact\ncomplementary. Combining the two produces predictions that look more realistic\nto human raters and better cover the range of possible futures. Our method\noutperforms prior and concurrent work in these aspects.","url_abs":"http://arxiv.org/abs/1804.01523v1","url_pdf":"http://arxiv.org/pdf/1804.01523v1.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":"stochastic-adversarial-video-prediction","repo_url":"https://github.com/alexlee-gk/video_prediction","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"stochastic-adversarial-video-prediction","repo_url":"https://github.com/Bonennult/video_prediction","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"stochastic-adversarial-video-prediction","repo_url":"https://github.com/MIT-Omnipush/video-prediction","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"stochastic-adversarial-video-prediction","repo_url":"https://github.com/kamran0153/impact-of-data-freshness-in-learning","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"prediction","task_name":"Prediction"},{"task_slug":"representation-learning","task_name":"Representation Learning"},{"task_slug":"video-generation","task_name":"Video Generation"},{"task_slug":"video-prediction","task_name":"Video Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/video-generation-on-bair-robot-pushing","task":"Video Generation","dataset":"BAIR Robot Pushing","model":"SAVP (from FVD)","rank_in_archive_order":12,"of":31,"metrics":{"Cond":"2","FVD score":"116.4","Pred":"14","Train":"14"},"uses_additional_data":false},{"leaderboard":"/sota/video-generation-on-bair-robot-pushing","task":"Video Generation","dataset":"BAIR Robot Pushing","model":"SAVP (from vRNN)","rank_in_archive_order":17,"of":31,"metrics":{"Cond":"2","FVD score":"143.43","LPIPS":"0.062±0.03","Pred":"28","SSIM":"0.795±0.07","Train":"10"},"uses_additional_data":false},{"leaderboard":"/sota/video-generation-on-bair-robot-pushing","task":"Video Generation","dataset":"BAIR Robot Pushing","model":"SAVP (from SRVP)","rank_in_archive_order":19,"of":31,"metrics":{"Cond":"2","FVD score":"152±9","LPIPS":"0.0634±0.0026","PSNR":"18.44±0.25","Pred":"28","SSIM":"0.7887±0.0092","Train":"12"},"uses_additional_data":false},{"leaderboard":"/sota/video-generation-on-bair-robot-pushing","task":"Video Generation","dataset":"BAIR Robot Pushing","model":"SAVP-VAE (from WAM)","rank_in_archive_order":31,"of":31,"metrics":{"Cond":"2","PSNR":"19.09","Pred":"28","SSIM":"0.815","Train":"14"},"uses_additional_data":false},{"leaderboard":"/sota/video-prediction-on-kth","task":"Video Prediction","dataset":"KTH","model":"SAVP-VAE (from Grid-keypoints)","rank_in_archive_order":2,"of":31,"metrics":{"Cond":"10","FVD":"145.7","LPIPS":"0.116","PSNR":"26.00","Params (M)":"7.3","Pred":"40","SSIM":"0.806","Train":"10"},"uses_additional_data":false},{"leaderboard":"/sota/video-prediction-on-kth","task":"Video Prediction","dataset":"KTH","model":"SAVP (from Grid-keypoints)","rank_in_archive_order":4,"of":31,"metrics":{"Cond":"10","FVD":"183.7","LPIPS":"0.126","PSNR":"23.79","Params (M)":"17.6","Pred":"40","SSIM":"0.699","Train":"10"},"uses_additional_data":false},{"leaderboard":"/sota/video-prediction-on-kth","task":"Video Prediction","dataset":"KTH","model":"SAVP (from SRVP)","rank_in_archive_order":9,"of":31,"metrics":{"Cond":"10","FVD":"374 ± 3","LPIPS":"0.1120±0.0039","PSNR":"26.51±0.29","Pred":"30","SSIM":"0.7564±0.0062","Train":"10"},"uses_additional_data":false},{"leaderboard":"/sota/video-prediction-on-kth","task":"Video Prediction","dataset":"KTH","model":"SAVP-VAE","rank_in_archive_order":19,"of":31,"metrics":{"Cond":"10","PSNR":"27.77","Pred":"20","SSIM":"0.852"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1804.01523","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1804.01523"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/kamran0153/impact-of-data-freshness-in-learning","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/alexlee-gk/video_prediction","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/MIT-Omnipush/video-prediction","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/Bonennult/video_prediction","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":2,"ran_fixture":1,"ran_violates":1,"unverified":12},"by_repo_kind":{"official":{"samples":2,"ran":2,"repositories":1},"listed":{"samples":13,"ran":2,"repositories":2}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":3,"samples":[{"code_sha256_prefix":"5815eb5ba521e8cb","entry":"load_metrics","repo":"alexlee-gk/video_prediction","repo_kind":"official","path":"scripts/evaluate.py","file_url":"https://github.com/alexlee-gk/video_prediction/blob/HEAD/scripts/evaluate.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5815eb5ba521e8cb"}},{"code_sha256_prefix":"7f866643ac8b4ba5","entry":"merge_hparams","repo":"alexlee-gk/video_prediction","repo_kind":"official","path":"scripts/evaluate.py","file_url":"https://github.com/alexlee-gk/video_prediction/blob/HEAD/scripts/evaluate.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"7f866643ac8b4ba5"}},{"code_sha256_prefix":"6beba365e8c55ed7","entry":"schedule_sampling","repo":"MIT-Omnipush/video-prediction","repo_kind":"listed","path":"svg/train_svg_lp.py","file_url":"https://github.com/MIT-Omnipush/video-prediction/blob/HEAD/svg/train_svg_lp.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"6beba365e8c55ed7"}},{"code_sha256_prefix":"c7792bb07554a763","entry":"should","repo":"MIT-Omnipush/video-prediction","repo_kind":"listed","path":"svg/train_svg_lp.py","file_url":"https://github.com/MIT-Omnipush/video-prediction/blob/HEAD/svg/train_svg_lp.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c7792bb07554a763"}},{"code_sha256_prefix":"a3b82c6547cef66d","entry":"bilinear_interp","repo":"Bonennult/video_prediction","repo_kind":"listed","path":"video_prediction/models/dvf_model.py","file_url":"https://github.com/Bonennult/video_prediction/blob/HEAD/video_prediction/models/dvf_model.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a3b82c6547cef66d"}},{"code_sha256_prefix":"dcd8505989cde89e","entry":"cem","repo":"Bonennult/video_prediction","repo_kind":"listed","path":"video_prediction/cem.py","file_url":"https://github.com/Bonennult/video_prediction/blob/HEAD/video_prediction/cem.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"dcd8505989cde89e"}},{"code_sha256_prefix":"0c7e34f278f8d941","entry":"charbonnier_loss","repo":"Bonennult/video_prediction","repo_kind":"listed","path":"video_prediction/losses.py","file_url":"https://github.com/Bonennult/video_prediction/blob/HEAD/video_prediction/losses.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0c7e34f278f8d941"}},{"code_sha256_prefix":"b0294c8d3f824240","entry":"foldl","repo":"Bonennult/video_prediction","repo_kind":"listed","path":"video_prediction/functional_ops.py","file_url":"https://github.com/Bonennult/video_prediction/blob/HEAD/video_prediction/functional_ops.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b0294c8d3f824240"}},{"code_sha256_prefix":"b85af650535732ac","entry":"image_warp","repo":"Bonennult/video_prediction","repo_kind":"listed","path":"video_prediction/flow_ops.py","file_url":"https://github.com/Bonennult/video_prediction/blob/HEAD/video_prediction/flow_ops.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b85af650535732ac"}},{"code_sha256_prefix":"1353b420a1f9ed3f","entry":"kl_criterion","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"1353b420a1f9ed3f"}},{"code_sha256_prefix":"8893afc8452805c9","entry":"l1_loss","repo":"Bonennult/video_prediction","repo_kind":"listed","path":"video_prediction/losses.py","file_url":"https://github.com/Bonennult/video_prediction/blob/HEAD/video_prediction/losses.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"8893afc8452805c9"}},{"code_sha256_prefix":"f9288c89552d4580","entry":"l2_loss","repo":"Bonennult/video_prediction","repo_kind":"listed","path":"video_prediction/losses.py","file_url":"https://github.com/Bonennult/video_prediction/blob/HEAD/video_prediction/losses.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"f9288c89552d4580"}},{"code_sha256_prefix":"537f14f6c82b799a","entry":"mean_squared_error_np","repo":"Bonennult/video_prediction","repo_kind":"listed","path":"video_prediction/metrics.py","file_url":"https://github.com/Bonennult/video_prediction/blob/HEAD/video_prediction/metrics.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"537f14f6c82b799a"}},{"code_sha256_prefix":"02fded9160d71360","entry":"noise","repo":"Bonennult/video_prediction","repo_kind":"listed","path":"video_prediction/models/mocogan_model.py","file_url":"https://github.com/Bonennult/video_prediction/blob/HEAD/video_prediction/models/mocogan_model.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"02fded9160d71360"}},{"code_sha256_prefix":"4c64121181ecb7d0","entry":"peak_signal_to_noise_ratio_np","repo":"Bonennult/video_prediction","repo_kind":"listed","path":"video_prediction/metrics.py","file_url":"https://github.com/Bonennult/video_prediction/blob/HEAD/video_prediction/metrics.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4c64121181ecb7d0"}},{"code_sha256_prefix":"e967d584d808f959","entry":"structural_similarity_np","repo":"Bonennult/video_prediction","repo_kind":"listed","path":"video_prediction/metrics.py","file_url":"https://github.com/Bonennult/video_prediction/blob/HEAD/video_prediction/metrics.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"e967d584d808f959"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}