wkentaro/pytorch-fcn

Health v2 · maintenance only

Flags: no_releases archived

Adoption not part of the score

1777 stars · 470 forks observed · 2026-08-28

What it is AI-extracted, prompt v1, taxonomy v1, 2026-08-30, confidence not recorded

A PyTorch implementation of Fully Convolutional Networks (FCN) for semantic segmentation, including training code that reproduces the original paper's results on PASCAL VOC. It provides FCN32s, FCN16s, FCN8s, and FCN8sAtOnce models with pretrained weights.

Use cases

• train a semantic segmentation model on pascal voc
• implement fully convolutional networks in pytorch
• download pretrained fcn8s weights for segmentation
• reproduce original fcn paper results
• pixel-wise image segmentation with deep learning
• compare fcn32s fcn16s fcn8s mean iou accuracy

When to choose

• you need the classic FCN architecture in PyTorch with verified reproduction of original results
• you want pretrained FCN models for PASCAL VOC segmentation
• you need a reference baseline for semantic segmentation research

When to avoid

• you need state-of-the-art segmentation models like DeepLab or Mask2Former
• you want a actively developed library with frequent updates
• you need segmentation for non-VOC datasets without writing your own training pipeline

Facets

library · maturity maintenance

deep-learning machine-learning image-processing computer-vision deep-learning machine-learning python semantic-segmentation fully-convolutional-networks pytorch fcn8s fcn32s fcn16s voc training-code gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="wkentaro/pytorch-fcn")