
Animated Drawings
Transform static character sketches into rigged 3D-mapped animations using Meta's FAIR computer vision architecture.
Pluggable SOTA multi-object tracking modules for segmentation, object detection, and pose estimation models.
BoxMOT is a Python package providing a modular architecture for multi-object tracking (MOT). It supports integration with various segmentation, object detection, and pose estimation models, enabling users to easily swap different SOTA tracking algorithms. The key value proposition lies in its pluggable architecture, universal model support, and benchmark-ready local evaluation pipelines for datasets like MOT17, MOT20, and DanceTrack. Performance modes include motion-only for lightweight CPU-efficient tracking and motion + appearance, combining motion cues with appearance embeddings (CLIPReID, LightMBN, OSNet) to maximize identity consistency and accuracy. It supports reusable detections and embeddings, which can be saved and reused for evaluations, eliminating redundant preprocessing. BoxMOT utilizes a command-line interface (CLI) for simplified syntax, allowing users to track objects, evaluate performance, tune hyperparameters, generate tracking data and export models.
BoxMOT is a Python package providing a modular architecture for multi-object tracking (MOT).
Explore all tools that specialize in pose estimation. This domain focus ensures BoxMOT delivers optimized results for this specific requirement.
Allows swapping different multi-object tracking algorithms (DeepOCSORT, BoTSORT, ByteTrack, StrongSORT, OCSort, HybridSORT, BoostTrack, SF-SORT) with minimal code changes.
Supports integration with any segmentation, object detection, and pose estimation models that output bounding boxes, providing flexibility in model selection.
Includes local evaluation pipelines for MOT17, MOT20, and DanceTrack datasets with 'official' ablation detectors.
Offers motion-only for lightweight, CPU-efficient, high-FPS performance and motion + appearance which combines motion cues with appearance embeddings for maximum identity consistency and accuracy.
Allows saving pre-generated detections and embeddings for reuse, eliminating redundant preprocessing steps.
Install Python 3.9 or higher.
Install the boxmot package using pip: `pip install boxmot`.
Set up your desired object detection or segmentation model (e.g., YOLOv8).
Download necessary ReID (Re-Identification) models (e.g., osnet_x0_25_msmt17.pt).
Use the CLI to specify the tracking mode, detector, ReID model, and tracker algorithm.
Run tracking on your desired video source or dataset.
All Set
Ready to go
Verified feedback from other users.
"BoxMOT offers robust multi-object tracking capabilities with a modular design and good performance."
Post questions, share tips, and help other users.

Transform static character sketches into rigged 3D-mapped animations using Meta's FAIR computer vision architecture.

The world's most advanced digital musculoskeletal clinic for personalized joint and muscle care.

Real-time object detection and image segmentation model optimized for edge deployment.