RefineNet semantic segmentation
Project description
~Please note this is only a beta release at this stage~
RefineNet: high-res semantic image segmentation
RefineNet is a generic multi-path refinement network for high-resolution semantic image segmentation and general dense prediction tasks on images. It achieves high-resolution prediction by explicitly exploiting all the information available along the down-sampling process and using long-range residual connections.
This repository contains an open-source implementation of RefineNet in Python, with both the official and lightweight network models from our publications. The package provides PyTorch implementations for training, evaluation, and deployment within systems. The package is easily installable with conda
, and can also be installed via pip
if you'd prefer to manually handle dependencies.
Our code is free to use, and licensed under BSD-3. We simply ask that you cite our work if you use RefineNet in your own research.
Related resources
This repository brings the work from a number of sources together. Please see the links below for further details:
- our original paper: "RefineNet: Multi-Path Refinement Networks for High-Resolution Semantic Segmentation"
- our paper introducing the lightweight version: "Light-Weight RefineNet for Real-Time Semantic Segmentation"
- the original MATLAB implementation: https://github.com/guosheng/refinenet
- Vladimir Nekrasov's PyTorch port of RefineNet: https://github.com/DrSleep/refinenet-pytorch
- Vladimir Nekrasov's PyTorch port of lightweight RefineNet: https://github.com/DrSleep/light-weight-refinenet
Installing RefineNet
We offer three methods for installing RefineNet:
- Through our Conda package: single command installs everything including system dependencies (recommended)
- Through our pip package: single command installs RefineNet and Python dependences, you take care of system dependencies
- Directly from source: allows easy editing and extension of our code, but you take care of building and all dependencies
Conda
The only requirement is that you have Conda installed on your system, and are inside a Conda environment. From there, simply run:
u@pc:~$ conda install refinenet
You can see a list of our Conda dependencies in the ./requirements.yml
file.
Pip
Before installing via pip
, you must have the following system dependencies installed:
- CUDA
- TODO the rest of this list
Then RefineNet, and all its Python dependencies can be installed via:
u@pc:~$ pip install refinenet
From source
Installing from source is very similar to the pip
method above due to RefineNet only containing Python code. Simply clone the repository, enter the directory, and install via pip
:
u@pc:~$ pip install -e .
Note: the editable mode flag (-e
) is optional, but allows you to immediately use any changes you make to the code in your local Python ecosystem.
Using RefineNet
Once installed, RefineNet can be used directly from the command line using Python
TODO: add details for quickstart scripts that run directly from the command line
Once installed, RefineNet can be used like any other Python package. It consists of a RefineNet
class with three main functions for training, evaluation, and deployment. Below are some examples to help get you started with RefineNet:
RefineNet API
The package also includes a full Python API that allows you to use RefineNet directly in your own projects.
TODO: documentation link?
The snippet below shows a number of examples of how to use RefineNet with your own projects:
from refinenet import RefineNet
# Initialise a full RefineNet network with no pre-trained model
r = RefineNet()
# Initialise a standard RefineNet network with a model pre-trained on NYU
r = RefineNet(model_type='full', load_pretrained='nyu')
# Initialise a lightweight RefineNet network with 40 classes
r = RefineNet(model='lightweight', num_classes=40)
# Load a previous snapshot from a 152 layer network
r = RefineNet(load_snapshot='/path/to/snapshot', num_resnet_layers=152)
# Train a new model on the NYU dataset with a custom learning rate
r.train('nyu', learning_rate=0.0005)
# Train a model with the adam optimiser & 8 workers, saving output to ~/output
r.train('voc', optimiser_type='adam', num_workers=8,
output_directory='~/output')
# Get a predicted segmentation as a NumPy image, given an input NumPy image
segmentation_image = r.predict(image=my_image)
# Save a segmentation image to file, given an image from another image file
r.predict(image_file='/my/prediction.jpg',
output_file='/my/segmentation/image.jpg')
# Evaluate your model's performance on the voc dataset, & save the results with
# images
r.eval('voc', output_directy='/my/results.json', output_images=True)
Citing our work
If using RefineNet in your work, please cite our original CVPR paper:
@InProceedings{Lin_2017_CVPR,
author = {Lin, Guosheng and Milan, Anton and Shen, Chunhua and Reid, Ian},
title = {RefineNet: Multi-Path Refinement Networks for High-Resolution Semantic Segmentation},
booktitle = {Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {July},
year = {2017}
}
Please also cite our BMVC paper on Light-Weight RefineNet if using the lightweight models:
@article{nekrasov2018light,
title={Light-weight refinenet for real-time semantic segmentation},
author={Nekrasov, Vladimir and Shen, Chunhua and Reid, Ian},
journal={arXiv preprint arXiv:1810.03272},
year={2018}
}
Project details
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
Hashes for refinenet-0.9.1-py3-none-any.whl
Algorithm | Hash digest | |
---|---|---|
SHA256 | b7acbf6862a152c7a3a64eecc72ca35ae6f463d324e511273c316845ce4e9961 |
|
MD5 | 8d185450ee4c8a4bbb43feec941db145 |
|
BLAKE2b-256 | 28bebf5c4cddc010bab543cea65ae4ff3d5908e280495af479b77e456529a9d7 |