boxmot and AB3DMOT
These two tools are competitors because BoxMOT offers a pluggable, state-of-the-art framework for 2D multi-object tracking with support for various bounding box types, while AB3DMOT focuses specifically on providing a baseline and evaluation metrics for 3D multi-object tracking.
About boxmot
mikel-brostrom/boxmot
BoxMOT: Pluggable SOTA multi-object tracking modules with support for axis-aligned and oriented bounding boxes
Combines pluggable tracker implementations (BoTSORT, ByteTrack, DeepOCSORT, etc.) with unified CLI and Python APIs for detection-agnostic multi-object tracking. Supports caching detections and embeddings across experiments to accelerate benchmarking, and natively handles both axis-aligned and oriented bounding box formats. Includes local evaluation workflows for MOT17, MOT20, and DanceTrack datasets with hyperparameter tuning and model export capabilities.
About AB3DMOT
xinshuoweng/AB3DMOT
(IROS 2020, ECCVW 2020) Official Python Implementation for "3D Multi-Object Tracking: A Baseline and New Evaluation Metrics"
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work