All Projects → zacario-li → Fast-SCNN_pytorch

zacario-li / Fast-SCNN_pytorch

Licence: MIT license
A PyTorch Implementation of Fast-SCNN: Fast Semantic Segmentation Network(PyTorch >= 1.4)

Programming Languages

python
139335 projects - #7 most used programming language

Projects that are alternatives of or similar to Fast-SCNN pytorch

Dilation Tensorflow
A native Tensorflow implementation of semantic segmentation according to Multi-Scale Context Aggregation by Dilated Convolutions (2016). Optionally uses the pretrained weights by the authors.
Stars: ✭ 134 (+346.67%)
Mutual labels:  segmentation, semantic-segmentation
wasr network
WaSR Segmentation Network for Unmanned Surface Vehicles v0.5
Stars: ✭ 32 (+6.67%)
Mutual labels:  segmentation, semantic-segmentation
Kiu Net Pytorch
Official Pytorch Code of KiU-Net for Image Segmentation - MICCAI 2020 (Oral)
Stars: ✭ 134 (+346.67%)
Mutual labels:  segmentation, semantic-segmentation
Crfasrnn pytorch
CRF-RNN PyTorch version http://crfasrnn.torr.vision
Stars: ✭ 102 (+240%)
Mutual labels:  segmentation, semantic-segmentation
Fcn
Chainer Implementation of Fully Convolutional Networks. (Training code to reproduce the original result is available.)
Stars: ✭ 211 (+603.33%)
Mutual labels:  segmentation, semantic-segmentation
Openvehiclevision
An opensource lib. for vehicle vision applications (written by MATLAB), lane marking detection, road segmentation
Stars: ✭ 120 (+300%)
Mutual labels:  segmentation, semantic-segmentation
Hrnet Semantic Segmentation
The OCR approach is rephrased as Segmentation Transformer: https://arxiv.org/abs/1909.11065. This is an official implementation of semantic segmentation for HRNet. https://arxiv.org/abs/1908.07919
Stars: ✭ 2,369 (+7796.67%)
Mutual labels:  segmentation, semantic-segmentation
Deep Segmentation
CNNs for semantic segmentation using Keras library
Stars: ✭ 69 (+130%)
Mutual labels:  segmentation, semantic-segmentation
Semantic Segmentation Suite
Semantic Segmentation Suite in TensorFlow. Implement, train, and test new Semantic Segmentation models easily!
Stars: ✭ 2,395 (+7883.33%)
Mutual labels:  segmentation, semantic-segmentation
Keras Unet
Helper package with multiple U-Net implementations in Keras as well as useful utility tools helpful when working with image semantic segmentation tasks. This library and underlying tools come from multiple projects I performed working on semantic segmentation tasks
Stars: ✭ 196 (+553.33%)
Mutual labels:  segmentation, semantic-segmentation
Segmentation
Tensorflow implementation : U-net and FCN with global convolution
Stars: ✭ 101 (+236.67%)
Mutual labels:  segmentation, semantic-segmentation
CAP augmentation
Cut and paste augmentation for object detection and instance segmentation
Stars: ✭ 93 (+210%)
Mutual labels:  segmentation, semantic-segmentation
Setr Pytorch
Rethinking Semantic Segmentation from a Sequence-to-Sequence Perspective with Transformers
Stars: ✭ 96 (+220%)
Mutual labels:  segmentation, semantic-segmentation
Vision4j Collection
Collection of computer vision models, ready to be included in a JVM project
Stars: ✭ 132 (+340%)
Mutual labels:  segmentation, semantic-segmentation
3dunet abdomen cascade
Stars: ✭ 91 (+203.33%)
Mutual labels:  segmentation, semantic-segmentation
Keras Segmentation Deeplab V3.1
An awesome semantic segmentation model that runs in real time
Stars: ✭ 156 (+420%)
Mutual labels:  segmentation, semantic-segmentation
Semanticsegmentation
A framework for training segmentation models in pytorch on labelme annotations with pretrained examples of skin, cat, and pizza topping segmentation
Stars: ✭ 52 (+73.33%)
Mutual labels:  segmentation, semantic-segmentation
Multiclass Semantic Segmentation Camvid
Tensorflow 2 implementation of complete pipeline for multiclass image semantic segmentation using UNet, SegNet and FCN32 architectures on Cambridge-driving Labeled Video Database (CamVid) dataset.
Stars: ✭ 67 (+123.33%)
Mutual labels:  segmentation, semantic-segmentation
Imgclsmob
Sandbox for training deep learning networks
Stars: ✭ 2,405 (+7916.67%)
Mutual labels:  segmentation, semantic-segmentation
DocuNet
Code and dataset for the IJCAI 2021 paper "Document-level Relation Extraction as Semantic Segmentation".
Stars: ✭ 84 (+180%)
Mutual labels:  segmentation, semantic-segmentation

Fast-SCNN

基于如下论文复现该网络

fast-scnn paper

网络架构

network arch

网络结构详情

Input Block t c n s
1024 × 2048 × 3 Conv2D - 32 1 2
512 × 1024 × 32 DSConv - 48 1 2
256 × 512 × 48 DSConv - 64 1 2
128 × 256 × 64 bottleneck 6 64 3 2
64 × 128 × 64 bottleneck 6 96 3 2
32 × 64 × 96 bottleneck 6 128 3 1
32 × 64 × 128 PPM - 128 - -
32 × 64 × 128 FFM - 128 - -
128 × 256 × 128 DSConv - 128 2 1
128 × 256 × 128 Conv2D - nums of classes 1 1

Table 1

Input Operator Output
h × w × c Conv2D 1/1, f h × w × tc
h × w × tc DWConv 3/s, f h/s x w/s x tc
h/s x w/s x tc Conv2D 1/1, − h/s x w/s x c'

Table 2

使用方法

基于540x540分辨率,在voc2012数据集上训练了一个权重,各位可以用这个来初始化,节约一些训练时间
https://pan.baidu.com/s/17_pGbpkI4tx8eOMZFS73fA password:v98k

数据准备

准备原图文件夹 img,准备label图文件夹 label,然后准备好train.txt 和 val.txt,放在同一级目录下,结构如下:
dataset
|train.txt
|val.txt
└--img
| | image1.jpg
| | image2.jpg
└--label
| image1.png
| image2.png

train.txt/val.txt 格式如下:
image1.jpg image1.png
image2.jpg image2.png
......

训练脚本准备

可以按照上述结构准备数据并训练

TODO

  • Training & Validate functions
  • Tensorboard 记录
  • resume training 脚本
  • VOC2012数据集训练脚本
  • 多GPU训练
Note that the project description data, including the texts, logos, images, and/or trademarks, for each open source project belongs to its rightful owner. If you wish to add or remove any projects, please contact us at [email protected].