All Projects → kostyaev → ICNR

kostyaev / ICNR

Licence: MIT license
Convolution NN resize initialization for subpixel convolutions

Programming Languages

Jupyter Notebook
11667 projects
python
139335 projects - #7 most used programming language

Projects that are alternatives of or similar to ICNR

FB
Multi-frame super-resolution via sub-pixel.
Stars: ✭ 55 (+77.42%)
Mutual labels:  subpixel
SwiftLoadHook
Use a hack way to achieve similar functions as Load() or initialize() in OC
Stars: ✭ 21 (-32.26%)
Mutual labels:  initializer
pytorch-gans
PyTorch implementation of GANs (Generative Adversarial Networks). DCGAN, Pix2Pix, CycleGAN, SRGAN
Stars: ✭ 21 (-32.26%)
Mutual labels:  superresolution
nxt init
Define initializers and attribute readers for your arguments with a single line of code
Stars: ✭ 21 (-32.26%)
Mutual labels:  initializer
redux-preloader
A Higher Order Component that helps to initialize data.
Stars: ✭ 14 (-54.84%)
Mutual labels:  initializer
libpointing
An open-source cross-platform library to get raw events from pointing devices and master transfer functions.
Stars: ✭ 45 (+45.16%)
Mutual labels:  subpixel
RCAN-tf
TensorFlow code for ECCV 2018 paper "Image Super-Resolution Using Very Deep Residual Channel Attention Networks"
Stars: ✭ 25 (-19.35%)
Mutual labels:  superresolution
Mobile Image-Video Enhancement
Sensifai image and video enhancement module on mobiles
Stars: ✭ 39 (+25.81%)
Mutual labels:  superresolution
superresolution gan
Chainer implementation of Photo-Realistic Single Image Super-Resolution Using a Generative Adversarial Network
Stars: ✭ 50 (+61.29%)
Mutual labels:  superresolution
SuperResolution-with-Perceptual-loss
In this project I have used an convolution neural network with perceptual loss to convert low res image into high res image.
Stars: ✭ 17 (-45.16%)
Mutual labels:  superresolution
TensorFlow-SR-DenseNet
Tensorflow with Image Super-Resolution Using Dense Skip Connections , color
Stars: ✭ 36 (+16.13%)
Mutual labels:  superresolution

Checkerboard artifact free sub pixel convolution initialization

Tensorflow implementation of ICNR initialization used in https://arxiv.org/pdf/1707.02937.pdf

screenshot

Updated for TF v2.4.

Usage example:

Wrap up your initialization with ICRN and you are ready to go:

layers.Conv2DTranspose(..., strides=scale, kernel_initializer=ICNR(GlorotUniform(), scale))
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].