FutrTrack: A Camera-LiDAR Fusion Transformer for 3D Multiple Object Tracking

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Teye, Martha Teiko, Maoz, Ori, Rottmann, Matthias
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866911321553895424
author Teye, Martha Teiko
Maoz, Ori
Rottmann, Matthias
author_facet Teye, Martha Teiko
Maoz, Ori
Rottmann, Matthias
contents We propose FutrTrack, a modular camera-LiDAR multi-object tracking framework that builds on existing 3D detectors by introducing a transformer-based smoother and a fusion-driven tracker. Inspired by query-based tracking frameworks, FutrTrack employs a multimodal two-stage transformer refinement and tracking pipeline. Our fusion tracker integrates bounding boxes with multimodal bird's-eye-view (BEV) fusion features from multiple cameras and LiDAR without the need for an explicit motion model. The tracker assigns and propagates identities across frames, leveraging both geometric and semantic cues for robust re-identification under occlusion and viewpoint changes. Prior to tracking, we refine sequences of bounding boxes with a temporal smoother over a moving window to refine trajectories, reduce jitter, and improve spatial consistency. Evaluated on nuScenes and KITTI, FutrTrack demonstrates that query-based transformer tracking methods benefit significantly from multimodal sensor features compared with previous single-sensor approaches. With an aMOTA of 74.7 on the nuScenes test set, FutrTrack achieves strong performance on 3D MOT benchmarks, reducing identity switches while maintaining competitive accuracy. Our approach provides an efficient framework for improving transformer-based trackers to compete with other neural-network-based methods even with limited data and without pretraining.
format Preprint
id arxiv_https___arxiv_org_abs_2510_19981
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FutrTrack: A Camera-LiDAR Fusion Transformer for 3D Multiple Object Tracking
Teye, Martha Teiko
Maoz, Ori
Rottmann, Matthias
Computer Vision and Pattern Recognition
We propose FutrTrack, a modular camera-LiDAR multi-object tracking framework that builds on existing 3D detectors by introducing a transformer-based smoother and a fusion-driven tracker. Inspired by query-based tracking frameworks, FutrTrack employs a multimodal two-stage transformer refinement and tracking pipeline. Our fusion tracker integrates bounding boxes with multimodal bird's-eye-view (BEV) fusion features from multiple cameras and LiDAR without the need for an explicit motion model. The tracker assigns and propagates identities across frames, leveraging both geometric and semantic cues for robust re-identification under occlusion and viewpoint changes. Prior to tracking, we refine sequences of bounding boxes with a temporal smoother over a moving window to refine trajectories, reduce jitter, and improve spatial consistency. Evaluated on nuScenes and KITTI, FutrTrack demonstrates that query-based transformer tracking methods benefit significantly from multimodal sensor features compared with previous single-sensor approaches. With an aMOTA of 74.7 on the nuScenes test set, FutrTrack achieves strong performance on 3D MOT benchmarks, reducing identity switches while maintaining competitive accuracy. Our approach provides an efficient framework for improving transformer-based trackers to compete with other neural-network-based methods even with limited data and without pretraining.
title FutrTrack: A Camera-LiDAR Fusion Transformer for 3D Multiple Object Tracking
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2510.19981