Track-to-Track Association for Collective Perception based on Stochastic Optimization
Collective perception is a key aspect for autonomous driving in smart cities as it aims to combine the local environment models of multiple intelligent vehicles in order to overcome sensor limitations. A crucial part of multi-sensor fusion is track-to-track association. Previous works often suffer from high computational complexity or are based on heuristics. We propose an association algorithms based on stochastic optimization, which leverages a multidimensional likelihood incorporating the number of tracks and their spatial distribution and furthermore computes several association hypotheses. We demonstrate the effectiveness of our approach in Monte Carlo simulations and a realistic collective perception scenario computing high-likelihood associations in ambiguous settings.
Code (0)
등록된 구현이 없습니다.
Tasks
Stochastic OptimizationAutonomous DrivingSimilar Papers 제목 키워드 기반
CoopTrack: Exploring End-to-End Learning for Efficient Cooperative Sequential Perception
Cooperative perception aims to address the inherent limitations of single-vehicle autonomous driving systems through information exchange among multiple agents. Previous research has primarily focused on single-frame per…
3D Multi-Object TrackingAutonomous DrivingVehicle Tracking in Wide Area Motion Imagery via Stochastic Progressive Association Across Multiple Frames (SPAAM)
Vehicle tracking in Wide Area Motion Imagery (WAMI) relies on associating vehicle detections across multiple WAMI frames to form tracks corresponding to individual vehicles. The temporal window length, i.e., the number $…
All Weather Perception: Joint Data Association, Tracking, and Classification for Autonomous Ground Vehicles
A novel probabilistic perception algorithm is presented as a real-time joint solution to data association, object tracking, and object classification for an autonomous ground vehicle in all-weather conditions. The presen…
AllClassificationGeneral ClassificationMulti-Object Tracking+2MOTR: End-to-End Multiple-Object Tracking with Transformer
Temporal modeling of objects is a key challenge in multiple object tracking (MOT). Existing methods track by associating detections through motion-based and appearance-based similarity heuristics. The post-processing nat…
Multi-Object TrackingMultiple Object TrackingMultiple Object Tracking with TransformerObject Detection+1OpenPifPaf: Composite Fields for Semantic Keypoint Detection and Spatio-Temporal Association
Many image-based perception tasks can be formulated as detecting, associating and tracking semantic keypoints, e.g., human body pose estimation and tracking. In this work, we present a general framework that jointly dete…
Car Pose EstimationKeypoint DetectionMulti-Person Pose EstimationPose Estimation+1