ZQPei/deep_sort_pytorch

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

3012 stars · 726 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-29, confidence not recorded

A PyTorch implementation of the Deep SORT multi-object tracking algorithm, pairing YOLOv3/YOLOv5 (or Mask R-CNN) detectors with a CNN re-identification feature extractor and Kalman filtering. It is a re-implementation of the original TensorFlow Deep SORT code.

Use cases

• track multiple objects in video with unique IDs
• pedestrian tracking with deep sort
• object detection and tracking with yolov3 in pytorch
• re-identify people across video frames
• count and track people in surveillance footage
• instance segmentation with tracking masks

When to choose

• you want a PyTorch alternative to the original TensorFlow Deep SORT
• you need detector-agnostic MOT with YOLOv3/v5 or Mask R-CNN
• you want to retrain the REID feature extractor yourself

When to avoid

• you need state-of-the-art trackers like ByteTrack or BoT-SORT
• you need a production-ready maintained pipeline rather than research code
• you work outside Python/PyTorch ecosystems

Facets

library · maturity active

computer-vision machine-learning deep-learning computer-vision deep-learning python object-tracking multi-object-tracking yolov3 yolov5 reid kalman-filter pytorch

Member repositories

For agents

markdown · JSON · MCP: product_card(name="ZQPei/deep_sort_pytorch")