{"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/desire-distant-future-prediction-in-dynamic","title":"DESIRE: Distant Future Prediction in Dynamic Scenes with Interacting Agents","arxiv_id":"1704.04394","date":"2017-04-14","proceeding":"CVPR 2017 7","authors":["Namhoon Lee","Wongun Choi","Paul Vernaza","Christopher B. Choy","Philip H. S. Torr","Manmohan Chandraker"],"abstract":"We introduce a Deep Stochastic IOC RNN Encoderdecoder framework, DESIRE, for\nthe task of future predictions of multiple interacting agents in dynamic\nscenes. DESIRE effectively predicts future locations of objects in multiple\nscenes by 1) accounting for the multi-modal nature of the future prediction\n(i.e., given the same context, future may vary), 2) foreseeing the potential\nfuture outcomes and make a strategic prediction based on that, and 3) reasoning\nnot only from the past motion history, but also from the scene context as well\nas the interactions among the agents. DESIRE achieves these in a single\nend-to-end trainable neural network model, while being computationally\nefficient. The model first obtains a diverse set of hypothetical future\nprediction samples employing a conditional variational autoencoder, which are\nranked and refined by the following RNN scoring-regression module. Samples are\nscored by accounting for accumulated future rewards, which enables better\nlong-term strategic decisions similar to IOC frameworks. An RNN scene context\nfusion module jointly captures past motion histories, the semantic scene\ncontext and interactions among multiple agents. A feedback mechanism iterates\nover the ranking and refinement to further boost the prediction accuracy. We\nevaluate our model on two publicly available datasets: KITTI and Stanford Drone\nDataset. Our experiments show that the proposed model significantly improves\nthe prediction accuracy compared to other baseline methods.","url_abs":"http://arxiv.org/abs/1704.04394v1","url_pdf":"http://arxiv.org/pdf/1704.04394v1.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":"desire-distant-future-prediction-in-dynamic","repo_url":"https://github.com/MiG98789/urop-desire","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"unanswered"}},{"paper_slug":"desire-distant-future-prediction-in-dynamic","repo_url":"https://github.com/AkashGanesan/desire-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":null},{"paper_slug":"desire-distant-future-prediction-in-dynamic","repo_url":"https://github.com/yadrimz/DESIRE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"future-prediction","task_name":"Future prediction"},{"task_slug":"multi-future-trajectory-prediction-1","task_name":"Multi Future Trajectory Prediction"},{"task_slug":"prediction","task_name":"Prediction"},{"task_slug":"trajectory-prediction","task_name":"Trajectory Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/trajectory-prediction-on-interaction-dataset-2","task":"Trajectory Prediction","dataset":"INTERACTION Dataset - Validation","model":"DESIRE","rank_in_archive_order":3,"of":4,"metrics":{"minADE6":"0.32","minFDE6":"0.88"},"uses_additional_data":false},{"leaderboard":"/sota/trajectory-prediction-on-paid","task":"Trajectory Prediction","dataset":"PAID","model":"DESIRE","rank_in_archive_order":1,"of":3,"metrics":{"minADE3":"0.29","minFDE3":"0.59"},"uses_additional_data":false},{"leaderboard":"/sota/trajectory-prediction-on-stanford-drone","task":"Trajectory Prediction","dataset":"Stanford Drone","model":"DESIRE","rank_in_archive_order":17,"of":24,"metrics":{"ADE (8/12) @K=5":"19.25","ADE-8/12 @K = 20":"19.25","FDE(8/12) @K=5":"34.05","FDE-8/12 @K= 20":"34.05"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1704.04394","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1704.04394"}},"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/AkashGanesan/desire-pytorch","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/MiG98789/urop-desire","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/yadrimz/DESIRE","reach":{"status":"unanswered"}}],"summary":{"ran_honours":1},"by_repo_kind":{"listed":{"samples":1,"ran":1,"repositories":1}},"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":1,"samples":[{"code_sha256_prefix":"c87e08e7b5f708e9","entry":"to_var","repo":"AkashGanesan/desire-pytorch","repo_kind":"listed","path":"desire/nn/cvae.py","file_url":"https://github.com/AkashGanesan/desire-pytorch/blob/HEAD/desire/nn/cvae.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c87e08e7b5f708e9"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}