#5454th of 56 in ML frameworks and libraries

paddledetection

Object detection toolkit on PaddlePaddle with tracking, keypoints and export tools

Stars
14.4k
License
Apache-2.0
Last commit
Mar 2026
Last release
Mar 2026
Language
Python

Missed the latest GitHub check; these facts may be older.

Overview

PaddleDetection is an end-to-end object detection suite built on PaddlePaddle. It bundles modular backbones, necks and losses with a model zoo covering 2D detection, instance segmentation, rotated boxes, multi-object tracking and keypoints, plus pipelines such as PP-Human and PP-Vehicle. Models export to Paddle Inference, Paddle Lite, Paddle Serving and ONNX.

Who it is for: CV engineers training and deploying detection models on PaddlePaddle

Strengths

  • Wide model zoo: Faster RCNN, YOLOX, PP-YOLOE, RT-DETR, PicoDet, DETR variants
  • Deploys via Paddle Inference (Python/C++), Paddle Lite, Paddle Serving or ONNX export
  • Ready-made pedestrian and vehicle analysis, tracking and keypoint pipelines
  • Training supports FP16, multi-machine runs, and PaddleSlim pruning, quantization, distillation

Weaknesses

  • Requires the PaddlePaddle framework rather than PyTorch
  • README and many docs are primarily in Chinese
  • README quick start now points to PaddleX, not the native workflow
  • No Docker or compose files detected in the repository

What it needs

  • GPU optional
  • Needs PaddlePaddle, PaddleSlim, Paddle Lite, Paddle Serving
  • Models: PP-YOLOE, RT-DETR, PP-PicoDet, Faster RCNN, YOLOX

Also in ML frameworks and libraries

See all 56
Also in ML frameworks and libraries
RankProjectScore
1transformersPython library defining open models for text, vision, audio and multimodal tasks167.7k starsApache-2.095 out of 100
2pytorchPython tensor library with GPU support and tape-based autograd for deep learning104.7k starscustom license94 out of 100
3ultralyticsPython library and CLI for training and running YOLO vision models62.5k starsAGPL-3.092 out of 100
#4supervisionPython toolkit for loading, annotating and counting computer vision detections Live demo ↗ (opens in a new tab)51.2k starsMIT89 out of 100
#5pytorch-image-modelsPyTorch library of image models, pretrained weights and reference training scripts37.2k starsApache-2.089 out of 100