{"type":"ScholarlyArticle","title":"BEVFormer++: Temporal Amplified BEVformer with Explicit Parameter Prediction for Automatic Trajectory Prediction","authors":[{"name":"Jiabin Fang","slug":"jiabin-fang","given":"Jiabin","family":"Fang","affiliation":""},{"name":"Xu Zhang","slug":"xu-zhang","given":"Xu","family":"Zhang","affiliation":""},{"name":"Zhuoming Ding","slug":"zhuoming-ding","given":"Zhuoming","family":"Ding","affiliation":""},{"name":"Xuan Liu","slug":"xuan-liu","given":"Xuan","family":"Liu","affiliation":""},{"name":"Meifang Zhang","slug":"meifang-zhang","given":"Meifang","family":"Zhang","affiliation":""},{"name":"Jin Yuan","slug":"jin-yuan","given":"Jin","family":"Yuan","affiliation":""},{"name":"Yuyi Wang","slug":"yuyi-wang","given":"Yuyi","family":"Wang","affiliation":""}],"abstract":"Vision-based trajectory prediction with BEV representations has achieved promising results, yet existing methods often suffer from limited temporal modeling and insufficient characterization of motion dynamics. To address these issues, we propose a temporally enhanced framework with explicit motion parameter prediction. Specifically, we introduce BEVFormer++, which leverages multi-view images and BEV features from multiple preceding timesteps to generate more robust BEV representations, along with BEV differential features to capture temporal variations. Moreover, we propose a motion-parameter-decoupled tracking module that explicitly estimates velocity, acceleration, and heading angle, providing informative motion cues for trajectory prediction. Extensive experimental results demonstrate that our method outperforms state-of-the-art approaches and can be seamlessly integrated into existing vision-based frameworks, consistently yielding performance improvements.","doi":"10.24963/ijcai.2026/121","url":"https://ijcai.org/proceedings/2026/121","pdf":"https://ijcai.org/proceedings/2026/0121.pdf","markdown":"https://ijcai.org/proceedings/2026/121.md","bibtex":"https://ijcai.org/proceedings/2026/bibtex/121","volume":{"slug":"2026","title":"Proceedings of the Thirty-Fifth International Joint Conference on Artificial Intelligence","short_title":"IJCAI 2026","url":"https://ijcai.org/proceedings/2026/","year":2026,"location":"Bremen, Germany","dates":"15-21 August 2026","editors":"Diego Calvanese","isbn":"978-1-956792-09-6","publisher":"International Joint Conferences on Artificial Intelligence Organization","doi_prefix":"10.24963/ijcai.2026"},"section":"Computer Vision","track":"Main Track","number":121,"pages":{"first":1079,"last":1087},"keywords":["Computer Vision: Motion and tracking","Computer Vision: Action and behavior recognition","Computer Vision: Multimodal learning"],"videos":[],"references":[],"license_note":"Copyright International Joint Conferences on Artificial Intelligence."}