File Download
There are no files associated with this item.
Links for fulltext
(May Require Subscription)
- Publisher Website: 10.1007/978-3-031-19839-7_37
- Scopus: eid_2-s2.0-85142748239
- WOS: WOS:000903760400037
- Find via
Supplementary
- Citations:
- Appears in Collections:
Conference Paper: SpOT: Spatiotemporal Modeling for 3D Object Tracking
Title | SpOT: Spatiotemporal Modeling for 3D Object Tracking |
---|---|
Authors | |
Keywords | 3D object detection 3D object tracking Autonomous driving LiDAR NuScenes Dataset point clouds |
Issue Date | 2022 |
Publisher | Springer Nature Switzerland |
Citation | 17th European Conference on Computer Vision (ECCV 2022), Tel Aviv, Israel, 23-27 October 2022. In Avidan, S, Brostow, G, Cissé, M, et al. (Eds.), Computer Vision – ECCV 2022: 17th European Conference, Tel Aviv, Israel, October 23-27, 2022, Proceedings, Part XXXVIII, p. 639-656. Cham: Springer Nature Switzerland, 2022 How to Cite? |
Abstract | 3D multi-object tracking aims to uniquely and consistently identify all mobile entities through time. Despite the rich spatiotemporal information available in this setting, current 3D tracking methods primarily rely on abstracted information and limited history, e.g. single-frame object bounding boxes. In this work, we develop a holistic representation of traffic scenes that leverages both spatial and temporal information of the actors in the scene. Specifically, we reformulate tracking as a spatiotemporal problem by representing tracked objects as sequences of time-stamped points and bounding boxes over a long temporal history. At each timestamp, we improve the location and motion estimates of our tracked objects through learned refinement over the full sequence of object history. By considering time and space jointly, our representation naturally encodes fundamental physical priors such as object permanence and consistency across time. Our spatiotemporal tracking framework achieves state-of-the-art performance on the Waymo and nuScenes benchmarks. |
Persistent Identifier | http://hdl.handle.net/10722/325585 |
ISBN | |
ISSN | 2023 SCImago Journal Rankings: 0.606 |
ISI Accession Number ID | |
Series/Report no. | Lecture Notes in Computer Science ; 13698 |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Stearns, Colton | - |
dc.contributor.author | Rempe, Davis | - |
dc.contributor.author | Li, Jie | - |
dc.contributor.author | Ambruş, Rareş | - |
dc.contributor.author | Zakharov, Sergey | - |
dc.contributor.author | Guizilini, Vitor | - |
dc.contributor.author | Yang, Yanchao | - |
dc.contributor.author | Guibas, Leonidas J. | - |
dc.date.accessioned | 2023-02-27T07:34:33Z | - |
dc.date.available | 2023-02-27T07:34:33Z | - |
dc.date.issued | 2022 | - |
dc.identifier.citation | 17th European Conference on Computer Vision (ECCV 2022), Tel Aviv, Israel, 23-27 October 2022. In Avidan, S, Brostow, G, Cissé, M, et al. (Eds.), Computer Vision – ECCV 2022: 17th European Conference, Tel Aviv, Israel, October 23-27, 2022, Proceedings, Part XXXVIII, p. 639-656. Cham: Springer Nature Switzerland, 2022 | - |
dc.identifier.isbn | 9783031198380 | - |
dc.identifier.issn | 0302-9743 | - |
dc.identifier.uri | http://hdl.handle.net/10722/325585 | - |
dc.description.abstract | 3D multi-object tracking aims to uniquely and consistently identify all mobile entities through time. Despite the rich spatiotemporal information available in this setting, current 3D tracking methods primarily rely on abstracted information and limited history, e.g. single-frame object bounding boxes. In this work, we develop a holistic representation of traffic scenes that leverages both spatial and temporal information of the actors in the scene. Specifically, we reformulate tracking as a spatiotemporal problem by representing tracked objects as sequences of time-stamped points and bounding boxes over a long temporal history. At each timestamp, we improve the location and motion estimates of our tracked objects through learned refinement over the full sequence of object history. By considering time and space jointly, our representation naturally encodes fundamental physical priors such as object permanence and consistency across time. Our spatiotemporal tracking framework achieves state-of-the-art performance on the Waymo and nuScenes benchmarks. | - |
dc.language | eng | - |
dc.publisher | Springer Nature Switzerland | - |
dc.relation.ispartof | Computer Vision – ECCV 2022: 17th European Conference, Tel Aviv, Israel, October 23-27, 2022, Proceedings, Part XXXVIII | - |
dc.relation.ispartofseries | Lecture Notes in Computer Science ; 13698 | - |
dc.subject | 3D object detection | - |
dc.subject | 3D object tracking | - |
dc.subject | Autonomous driving | - |
dc.subject | LiDAR | - |
dc.subject | NuScenes Dataset | - |
dc.subject | point clouds | - |
dc.title | SpOT: Spatiotemporal Modeling for 3D Object Tracking | - |
dc.type | Conference_Paper | - |
dc.description.nature | link_to_subscribed_fulltext | - |
dc.identifier.doi | 10.1007/978-3-031-19839-7_37 | - |
dc.identifier.scopus | eid_2-s2.0-85142748239 | - |
dc.identifier.spage | 639 | - |
dc.identifier.epage | 656 | - |
dc.identifier.eissn | 1611-3349 | - |
dc.identifier.isi | WOS:000903760400037 | - |
dc.publisher.place | Cham | - |