Multi-Object Tracking

ROS2 Node for tracking 3D objects detected around the car.

Overview

This tracker uses a modified version of ByteTrack to track 3D bounding boxes. Core logic draws inspiration from ByteTrackV2 (Zhang et al., 2023). The implementation is built on top of ByteTrack-cpp with modifications made for 3D tracking.

Topics

Subscribed

Published

Parameters

| Parameter | Type | Default | Description | |———–|——|———|————-| | frame_rate| int | 30 | Frame rate of the input detections | | track_buffer | int | 30 | Amount of consecutive frames a track can stay unmatched before getting removed | | track_thresh | float | 0.5 | Threshold between high and low confidence detections | | high_thresh | float | 0.6 | Minimum detection confidence score required to start a new track | | match_thresh | float | 0.8 | Maximum IoU cost to still be considered a match | | use_maj_cls | bool | true | Use most frequent class as track’s class if true, use most recent class otherwise | | output_frame | string | “map” | Frame to output tracks in |

Acknowledgements

This package uses a modified version of ByteTrack-cpp. The original repository is licensed under MIT License (see THIRD_PARTY_LICENSES/BYTETRACK_LICENSE for details).

ByteTrackV2 paper can be found at:
Y. Zhang et al., “ByteTrackV2: 2D and 3D Multi-Object Tracking by Associating Every Detection Box,” arXiv, Mar. 2023, https://doi.org/10.48550/arXiv.2303.15334