Movatterモバイル変換

[0]ホーム
URL:

Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly

 Sign up
Appearance settings

BoxMOT: Pluggable SOTA multi-object tracking modules modules for segmentation, object detection and pose estimation models

License

NotificationsYou must be signed in to change notification settings

mikel-brostrom/boxmot

Repository files navigation

BoxMot demo

mikel-brostrom%2Fboxmot | Trendshift

CIPyPI versiondownloadslicensepython-versioncolabDOIdocker pullsdiscordAsk DeepWiki

🚀 Key Features

  • Pluggable Architecture
    Easily swap in/out SOTA multi-object trackers.

  • Universal Model Support
    Integrate with any segmentation, object-detection and pose-estimation models that outputs bounding boxes

  • Benchmark-Ready
    Local evaluation pipelines for MOT17, MOT20, and DanceTrack ablation datasets with "official" ablation detectors

  • Performance Modes

    • Motion-only: for lightweight, CPU-efficient, high-FPS performance
    • Motion + Appearance: Combines motion cues with appearance embeddings (CLIPReID,LightMBN,OSNet) to maximize identity consistency and accuracy at a higher computational cost

  • Reusable Detections & Embeddings
    Save once, run evaluations with no redundant preprocessing lightning fast.

📊 Benchmark Results (MOT17 ablation split)

TrackerStatusHOTA↑MOTA↑IDF1↑FPS
boosttrack✅69.25375.91483.20625
botsort✅68.88578.22281.34446
hybridsort✅68.21676.38281.16425
strongsort✅68.0576.18580.76317
deepocsort✅67.79675.86880.51412
bytetrack✅67.6878.03979.1571265
ocsort✅66.44174.54877.8991483

NOTES: Evaluation was conducted on the second half of the MOT17 training set, as the validation set is not publicly available and the ablation detector was trained on the first half. We employedpre-generated detections and embeddings. Each tracker was configured using the default parameters from their official repositories.

🔧 Installation

Install theboxmot package, including all requirements, in a Python>=3.9 environment:

pip install boxmot

If you want to contribute to this package check how to contributehere

💻 CLI

BoxMOT provides a unified CLIboxmot with the following subcommands:

Usage: boxmot COMMAND [ARGS]...Commands:eval      Evaluate tracking performanceexport    Export ReID models  generate  Generate detections and embeddings  track     Run tracking only  tune      Tune models via evolutionary algorithms

🐍 PYTHON

Seamlessly integrate BoxMOT directly into your Python MOT applications with your custom model.

importcv2importtorchimportnumpyasnpfrompathlibimportPathfromboxmotimportBoostTrackfromtorchvision.models.detectionimport (fasterrcnn_resnet50_fpn_v2,FasterRCNN_ResNet50_FPN_V2_WeightsasWeights)# Set devicedevice=torch.device('cuda'iftorch.cuda.is_available()else'cpu')# Load detector with pretrained weights and preprocessing transformsweights=Weights.DEFAULTdetector=fasterrcnn_resnet50_fpn_v2(weights=weights,box_score_thresh=0.5)detector.to(device).eval()transform=weights.transforms()# Initialize trackertracker=BoostTrack(reid_weights=Path('osnet_x0_25_msmt17.pt'),device=device,half=False)# Start video capturecap=cv2.VideoCapture(0)withtorch.inference_mode():whileTrue:success,frame=cap.read()ifnotsuccess:break# Convert frame to RGB and prepare for detectorrgb=cv2.cvtColor(frame,cv2.COLOR_BGR2RGB)tensor=torch.from_numpy(rgb).permute(2,0,1).to(torch.uint8)input_tensor=transform(tensor).to(device)# Run detectionoutput=detector([input_tensor])[0]scores=output['scores'].cpu().numpy()keep=scores>=0.5# Prepare detections for trackingboxes=output['boxes'][keep].cpu().numpy()labels=output['labels'][keep].cpu().numpy()filtered_scores=scores[keep]detections=np.concatenate([boxes,filtered_scores[:,None],labels[:,None]],axis=1)# Update tracker and draw results#   INPUT:  M X (x, y, x, y, conf, cls)#   OUTPUT: M X (x, y, x, y, id, conf, cls, ind)res=tracker.update(detections,frame)tracker.plot_results(frame,show_trajectories=True)# Show outputcv2.imshow('BoXMOT + Torchvision',frame)ifcv2.waitKey(1)&0xFF==ord('q'):break# Clean upcap.release()cv2.destroyAllWindows()

📝 Code Examples & Tutorials

Tracking
$ boxmot track --yolo-model rf-detr-base.pt# bboxes only  boxmot track --yolo-model yolox_s.pt# bboxes only  boxmot track --yolo-model yolo12n.pt# bboxes only  boxmot track --yolo-model yolo11n.pt# bboxes only  boxmot track --yolo-model yolov10n.pt# bboxes only  boxmot track --yolo-model yolov9c.pt# bboxes only  boxmot track --yolo-model yolov8n.pt# bboxes only                            yolov8n-seg.pt# bboxes + segmentation masks                            yolov8n-pose.pt# bboxes + pose estimation
Tracking methods
$ boxmot track --tracking-method deepocsort                                 strongsort                                 ocsort                                 bytetrack                                 botsort                                 boosttrack
Tracking sources

Tracking can be run on most video formats

$ boxmot track --source 0# webcam                        img.jpg# image                        vid.mp4# video                        path/# directory                        path/*.jpg# glob'https://youtu.be/Zgi9g1ksQHc'# YouTube'rtsp://example.com/media.mp4'# RTSP, RTMP, HTTP stream
Select ReID model

Some tracking methods combine appearance description and motion in the process of tracking. For those which use appearance, you can choose a ReID model based on your needs from thisReID model zoo. These model can be further optimized for you needs by thereid_export.py script

$ boxmot track --source 0 --reid-model lmbn_n_cuhk03_d.pt# lightweight                                       osnet_x0_25_market1501.pt                                       mobilenetv2_x1_4_msmt17.engine                                       resnet50_msmt17.onnx                                       osnet_x1_0_msmt17.pt                                       clip_market1501.pt# heavy                                       clip_vehicleid.pt                                      ...
Filter tracked classes

By default the tracker tracks all MS COCO classes.

If you want to track a subset of the classes that you model predicts, add their corresponding index after the classes flag,

boxmot track --source 0 --yolo-model yolov8s.pt --classes 16 17# COCO yolov8 model. Track cats and dogs, only

Here is a list of all the possible objects that a Yolov8 model trained on MS COCO can detect. Notice that the indexing for the classes in this repo starts at zero

Evaluation

Evaluate a combination of detector, tracking method and ReID model on standard MOT dataset or you custom one by

# reproduce MOT17 README results$ boxmoteval --yolo-model yolox_x_MOT17_ablation.pt --reid-model lmbn_n_duke.pt --tracking-method boosttrack --source MOT17-ablation --verbose# MOT20 results$ boxmoteval --yolo-model yolox_x_MOT20_ablation.pt --reid-model lmbn_n_duke.pt --tracking-method boosttrack --source MOT20-ablation --verbose# Dancetrack results$ boxmoteval --yolo-model yolox_x_dancetrack_ablation.pt --reid-model lmbn_n_duke.pt --tracking-method boosttrack --source dancetrack-ablation --verbose# metrics on custom dataset$ boxmoteval --yolo-model yolov8n.pt --reid-model osnet_x0_25_msmt17.pt --tracking-method deepocsort  --source ./assets/MOT17-mini/train --verbose

add--gsi to your command for postprocessing the MOT results by gaussian smoothed interpolation. Detections and embeddings are stored for the selected YOLO and ReID model respectively. They can then be loaded into any tracking algorithm. Avoiding the overhead of repeatedly generating this data.

Evolution

We use a fast and elitist multiobjective genetic algorithm for tracker hyperparameter tuning. By default the objectives are: HOTA, MOTA, IDF1. Run it by

# saves dets and embs under ./runs/dets_n_embs separately for each selected yolo and reid model$ boxmot generate --source ./assets/MOT17-mini/train --yolo-model yolov8n.pt yolov8s.pt --reid-model weights/osnet_x0_25_msmt17.pt# evolve parameters for specified tracking method using the selected detections and embeddings generated in the previous step$ boxmot tune --yolo-model yolov8n.pt --reid-model osnet_x0_25_msmt17.pt --n-trials 9 --tracking-method botsort --source ./assets/MOT17-mini/train

The set of hyperparameters leading to the best HOTA result are written to the tracker's config file.

Export

We support ReID model export to ONNX, OpenVINO, TorchScript and TensorRT

# export to ONNX$ boxmotexport --weights weights/osnet_x0_25_msmt17.pt --include onnx --device cpu# export to OpenVINO$ boxmotexport --weights weights/osnet_x0_25_msmt17.pt --include openvino --device cpu# export to TensorRT with dynamic input$ boxmotexport --weights weights/osnet_x0_25_msmt17.pt --include engine --device 0 --dynamic
Example DescriptionNotebook
Torchvision bounding box tracking with BoxMOTNotebook
Torchvision pose tracking with BoxMOTNotebook
Torchvision segmentation tracking with BoxMOTNotebook

Contributors

Contact

For BoxMOT bugs and feature requests please visitGitHub Issues.For business inquiries or professional support requests please send an email to:box-mot@outlook.com

Sponsor this project

    Packages

    No packages published

    Contributors44

    [8]ページ先頭
    ©2009-2025 Movatter.jp