YuwenXiong/py-R-FCN

Health v2 · maintenance only

Flags: no_releases

Adoption not part of the score

1043 stars · 459 forks observed · 2026-08-28

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

A Python implementation of R-FCN (Region-based Fully Convolutional Networks) for object detection, modified from the official MATLAB code and py-faster-rcnn. It supports joint training and alternative optimization with ResNet backbones, class-agnostic bounding box regression, and OHEM.

Use cases

• train an R-FCN object detection model on Pascal VOC
• run region-based fully convolutional network detection with ResNet-50 or ResNet-101
• reproduce R-FCN paper results with joint training
• experiment with OHEM hard example mining for detection
• compare class-agnostic vs class-aware bounding box regression

When to choose

• you need a Python R-FCN implementation with joint training support
• you are reproducing the R-FCN paper on VOC-style datasets
• you want to study or extend a classic two-stage object detector built on Caffe

When to avoid

• you need CPU-only training or inference - it is unsupported
• you need multi-GPU training - the MXNet Deformable R-FCN version is recommended instead
• you need a modern, actively maintained detector - the project is largely historical

Facets

library · maturity maintenance

machine-learning deep-learning computer-vision image-processing computer-vision deep-learning machine-learning python object-detection r-fcn caffe resnet pascal-voc ohem joint-training linux gpu

Member repositories

For agents

markdown · JSON · MCP: product_card(name="YuwenXiong/py-R-FCN")