{"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/infer-intermediate-representations-for-future","title":"INFER: INtermediate representations for FuturE pRediction","arxiv_id":"1903.10641","date":"2019-03-26","proceeding":null,"authors":["Shashank Srikanth","Junaid Ahmed Ansari","Karnik Ram R","Sarthak Sharma","Krishna Murthy J.","Madhava Krishna K"],"abstract":"In urban driving scenarios, forecasting future trajectories of surrounding\nvehicles is of paramount importance. While several approaches for the problem\nhave been proposed, the best-performing ones tend to require extremely detailed\ninput representations (eg. image sequences). But, such methods do not\ngeneralize to datasets they have not been trained on. We propose intermediate\nrepresentations that are particularly well-suited for future prediction. As\nopposed to using texture (color) information, we rely on semantics and train an\nautoregressive model to accurately predict future trajectories of traffic\nparticipants (vehicles) (see fig. above). We demonstrate that using semantics\nprovides a significant boost over techniques that operate over raw pixel\nintensities/disparities. Uncharacteristic of state-of-the-art approaches, our\nrepresentations and models generalize to completely different datasets,\ncollected across several cities, and also across countries where people drive\non opposite sides of the road (left-handed vs right-handed driving).\nAdditionally, we demonstrate an application of our approach in multi-object\ntracking (data association). To foster further research in transferrable\nrepresentations and ensure reproducibility, we release all our code and data.","url_abs":"http://arxiv.org/abs/1903.10641v1","url_pdf":"http://arxiv.org/pdf/1903.10641v1.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":"infer-intermediate-representations-for-future","repo_url":"https://github.com/talsperre/INFER","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"activity-prediction","task_name":"Activity Prediction"},{"task_slug":"future-prediction","task_name":"Future prediction"},{"task_slug":"multi-object-tracking","task_name":"Multi-Object Tracking"},{"task_slug":"object-tracking","task_name":"Object Tracking"},{"task_slug":"prediction","task_name":"Prediction"},{"task_slug":"trajectory-prediction","task_name":"Trajectory Prediction"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1903.10641","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}