#105105th of 105 in ML frameworks and libraries
paddleseg
Image segmentation toolkit on PaddlePaddle, from training to deployment
- Stars
- 9.4k
- License
- Apache-2.0
- Last commit
- Nov 2025
- Last release
- Nov 2024
- Language
- Python
9,253 on 1 Apr 20269,413 on 21 Oct 2026
From GitHub's daily star counts.
Checked on GitHub.
Overview
PaddleSeg is a PaddlePaddle-based toolkit for image segmentation with 45+ model algorithms and 140+ pretrained models. It covers semantic, interactive, matting and panoptic segmentation, driven by YAML configs or a Python API, and includes compression and deployment tooling. The README also points to PaddleX for low-code use of 19 segmentation and anomaly-detection models.
Who it is for: CV engineers training and deploying segmentation models on PaddlePaddle
Strengths
- Large model zoo: 45+ segmentation networks with Cityscapes and ADE20K benchmark tables
- Covers semantic, interactive (EISeg), matting and panoptic segmentation in one repo
- Includes quantization-aware distillation and deployment paths such as PaddleInference and PaddleLite
- Modular configs separate backbones, losses, datasets and augmentations
Weaknesses
- Requires the PaddlePaddle framework, not PyTorch, which limits reuse with PyTorch tooling
- Primary README is in Chinese, with English in a separate file
- Last release was 2024-11-05; no Docker or compose files detected
- Hardware and RAM requirements are not stated in the README excerpt
What it needs
- GPU optional
- Needs PaddlePaddle
- Models: PP-LiteSeg, PP-MobileSeg, DeepLabV3P, OCRNet, SegFormer
Also in ML frameworks and libraries
See all 105| Rank | Project | Score |
|---|---|---|
| 1 | transformersPython library defining open models for text, vision, audio and multimodal tasks | 94 out of 100 |
| 2 | pytorchPython tensor library with GPU support and tape-based autograd for deep learning | 94 out of 100 |
| 3 | ultralyticsPython library and CLI for training and running YOLO vision models | 93 out of 100 |
| #4 | pytorch-image-modelsPyTorch library of image models, pretrained weights and reference training scripts | 92 out of 100 |
| #5 | supervisionPython toolkit for loading, annotating and counting computer vision detections | 91 out of 100 |