All Projects → trevor-m → tensorflow-bicubic-downsample

trevor-m / tensorflow-bicubic-downsample

Licence: MIT license
tf.image.resize_images has aliasing when downsampling and does not have gradients for bicubic mode. This implementation fixes those problems.

Programming Languages

python
139335 projects - #7 most used programming language

Projects that are alternatives of or similar to tensorflow-bicubic-downsample

Bicubic-interpolation
Bicubic interpolation for images (Python)
Stars: ✭ 88 (+282.61%)
Mutual labels:  images, resize-images, super-resolution, image-resizing, bicubic
Lilliput
Resize images and animated GIFs in Go
Stars: ✭ 1,690 (+7247.83%)
Mutual labels:  images, resize-images
pixel
A lightweight image loader for Android backed by Kotlin Coroutines.
Stars: ✭ 79 (+243.48%)
Mutual labels:  images
pytorch-gans
PyTorch implementation of GANs (Generative Adversarial Networks). DCGAN, Pix2Pix, CycleGAN, SRGAN
Stars: ✭ 21 (-8.7%)
Mutual labels:  super-resolution
multer-sharp
Streaming multer storage engine permit to resize and upload to Google Cloud Storage.
Stars: ✭ 21 (-8.7%)
Mutual labels:  resize-images
pretty-formula
A small Java library to parse mathematical formulas to LaTeX and display them as images
Stars: ✭ 29 (+26.09%)
Mutual labels:  images
capella
Cloud service for image storage and delivery
Stars: ✭ 116 (+404.35%)
Mutual labels:  resize-images
RAImagePicker
📸 iMessage-like, Image Picker Controller Provides custom features.
Stars: ✭ 14 (-39.13%)
Mutual labels:  images
hermitage
Service that provides storage, delivery and modification of your images
Stars: ✭ 33 (+43.48%)
Mutual labels:  images
blur-up
A tool that creates preview images.
Stars: ✭ 28 (+21.74%)
Mutual labels:  images
ngx-fire-uploader
Angular Fire Uploader
Stars: ✭ 18 (-21.74%)
Mutual labels:  images
SRDenseNet-pytorch
SRDenseNet-pytorch(ICCV_2017)
Stars: ✭ 113 (+391.3%)
Mutual labels:  super-resolution
ESPCN-PyTorch
A PyTorch implementation of ESPCN based on CVPR 2016 paper Real-Time Single Image and Video Super-Resolution Using an Efficient Sub-Pixel Convolutional Neural Network.
Stars: ✭ 33 (+43.48%)
Mutual labels:  super-resolution
MSG-Net
Depth Map Super-Resolution by Deep Multi-Scale Guidance, ECCV 2016
Stars: ✭ 76 (+230.43%)
Mutual labels:  super-resolution
cdn
🚀 ✈️ 🚄 free CDN for everyone who wants to speed his website freely!😄
Stars: ✭ 16 (-30.43%)
Mutual labels:  images
tools-generation-detection-synthetic-content
Compilation of the state of the art of tools, articles, forums and links of interest to generate and detect any type of synthetic content using deep learning.
Stars: ✭ 107 (+365.22%)
Mutual labels:  images
danbooru-ruby-grabber
Danbooru, Konachan, Behoimi (3dbooru) and Yandere images downloader
Stars: ✭ 39 (+69.57%)
Mutual labels:  images
Everybody-dance-now
Implementation of paper everybody dance now for Deep learning course project
Stars: ✭ 22 (-4.35%)
Mutual labels:  images
seam-carving
A fast Python implementation of Seam Carving for Content-Aware Image Resizing.
Stars: ✭ 182 (+691.3%)
Mutual labels:  image-resizing
seam-carving
An implementation of the seam carving algorithm, designed to resize images while preventing distortion.
Stars: ✭ 30 (+30.43%)
Mutual labels:  image-resizing

tensorflow-bicubic-downsample

tf.image.resize_images has aliasing when downsampling and does not define gradients for bicubic mode. This implementation fixes those problems.

Example

These images have been downsampled by a factor of 4 from the original. The results from this code matches the scipy.misc.imresize results exactly.

Method Result Comments
Original This is the original full res image.
tf.images.resize_images TF's implementation has aliasing
scipy.misc.imresize Proper bicubic downsampling
This code Matches scipy exactly

Usage

from bicubic_downsample import build_filter, apply_bicubic_downsample

# First, create the bicubic kernel. This can be reused in multiple downsample operations
k = build_filter(factor=4)

# Downsample x which is a tensor with shape [N, H, W, 3]
y = apply_bicubic_downsample(x, filter=k, factor=4)

# y now contains x downsampled to [N, H/4, W/4, 3]
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].