ImageAI - A python library built to empower developers to build applications and systems with self-contained Computer Vision capabilities

•    Python

A python library built to empower developers to build applications and systems with self-contained Deep Learning and Computer Vision capabilities using simple and few lines of code. Built with simplicity in mind, ImageAI supports a list of state-of-the-art Machine Learning algorithms for image prediction, custom image prediction, object detection, video detection, video object tracking and image predictions trainings. ImageAI currently supports image prediction and training using 4 different Machine Learning algorithms trained on the ImageNet-1000 dataset. ImageAI also supports object detection, video detection and object tracking using RetinaNet, YOLOv3 and TinyYOLOv3 trained on COCO dataset. Eventually, ImageAI will provide support for a wider and more specialized aspects of Computer Vision including and not limited to image recognition in special environments and special fields.

tensornets - High level network definitions with pre-trained weights in TensorFlow

•    Python

High level network definitions with pre-trained weights in TensorFlow (tested with >= 1.1.0). You can install TensorNets from PyPI (pip install tensornets) or directly from GitHub (pip install git+https://github.com/taehoonlee/tensornets.git).

DenseNet - DenseNet implementation in Keras

•    Python

The Bottleneck - Compressed DenseNets offer further performance benefits, such as reduced number of parameters, with similar or better performance. The best original model, DenseNet-100-24 (27.2 million parameters) achieves 3.74 % error, whereas the DenseNet-BC-190-40 (25.6 million parameters) achieves 3.46 % error which is a new state of the art performance on CIFAR-10.

efficient_densenet_pytorch - A memory-efficient implementation of DenseNets

•    Python

A PyTorch implementation of DenseNets, optimized to save GPU memory. While DenseNets are fairly easy to implement in deep learning frameworks, most implmementations (such as the original) tend to be memory-hungry. In particular, the number of intermediate feature maps generated by batch normalization and concatenation operations grows quadratically with network depth. It is worth emphasizing that this is not a property inherent to DenseNets, but rather to the implementation.

densenet.pytorch - A PyTorch implementation of DenseNet.

•    Python

This is a PyTorch implementation of the DenseNet-BC architecture as described in the paper Densely Connected Convolutional Networks by G. Huang, Z. Liu, K. Weinberger, and L. van der Maaten. This implementation gets a CIFAR-10+ error rate of 4.77 with a 100-layer DenseNet-BC with a growth rate of 12. Their official implementation and links to many other third-party implementations are available in the liuzhuang13/DenseNet repo on GitHub. As this table from the DenseNet paper shows, it provides competitive state of the art results on CIFAR-10, CIFAR-100, and SVHN.

DenseNet-Caffe - DenseNet Caffe Models, converted from https://github.com/liuzhuang13/DenseNet

We manually converted the original torch models into caffe format from https://github.com/liuzhuang13/DenseNet. Update (July 27, 2017): for your convenience, we also provide a link to these models on Baidu Disk.

awesome-very-deep-learning - 🔥A curated list of papers and code about very deep neural networks

awesome-very-deep-learning is a curated list for papers and code about implementing and training very deep neural networks. Value Iteration Networks are very deep networks that have tied weights and perform approximate value iteration. They are used as an internal (model-based) planning module.

Densenet-Tensorflow - Simple Tensorflow implementation of "Densenet" using Cifar10, MNIST

•    Python

Simple Tensorflow implementation of "Densenet" using Cifar10, MNIST

SENet-Tensorflow - Simple Tensorflow implementation of "Squeeze and Excitation Networks" using Cifar10 (ResNeXt, Inception-v4, Inception-resnet-v2)

•    Python

Simple Tensorflow implementation of "Squeeze and Excitation Networks" using Cifar10 (ResNeXt, Inception-v4, Inception-resnet-v2)

pytorch-speech-commands - Speech commands recognition with PyTorch

•    Python

Convolutional neural networks for Google speech commands data set with PyTorch. We, xuyuan and tugstugi, have participated in the Kaggle competition TensorFlow Speech Recognition Challenge and reached the 10-th place. This repository contains a simplified and cleaned up version of our team's code.

chainer-cifar10 - Various CNN models including for CIFAR10 with Chainer

•    Python

See the details at transform function in train.py.

cnn-watermark-removal - Fully convolutional deep neural network to remove transparent overlays from images

•    Python

Deep learning architecture to remove transparent overlays from images. Bottom: Pascal dataset image reconstructions. When the watermarked area is saturated, the reconstruction tends to produce a gray color.

densenet - A PyTorch Implementation of "Densely Connected Convolutional Networks"

•    Python

This is a PyTorch implementation of the DenseNet architecture as described in Densely Connected Convolutional Networks by G. Huang, Z. Liu, K. Weinberger, and L. van der Maaten. This implementation currently supports training on the CIFAR-10 and CIFAR-100 datasets (support for ImageNet coming soon).

3ddensenet.torch - 3D DenseNet(torch version) for ModelNet40 dataset

•    Lua

See the installation instructions for a step-by-step guide. See the training recipes for addition examples.

DenseNet-MURA-PyTorch - Implementation of DenseNet model on Standford's MURA dataset using PyTorch

•    Python

The model takes as input one or more views for a study of an upper extremity. On each view, our 169-layer convolutional neural network predicts the probability of abnormality. We compute the overall probability of abnormality for the study by taking the arithmetic mean of the abnormality probabilities output by the network for each image. The model implemented in model.py takes as input 'all' the views for a study of an upper extremity. On each view the model predicts the probability of abnormality. The Model computes the overall probability of abnormality for the study by taking the arithmetic mean of the abnormality probabilites output by the network for each image.

Model-Playgrounds - A project developed and maintained as part of the aim at bringing current capabilities in machine learning and artificial intelligence into practical use for non-programmers and average computer users

•    Python

        The Machine Learning Model Playgrounds is a project that is part of the dream of a team of Moses Olafenwa and John Olafenwa to bring current capabilities in machine learning and artificial intelligence into practical use for non-programmers and average computer users. This project is the first step in what we hope will become mainstream application in modern technology in which Computers, Smartphones, Edge Devices and Systems will have in-built state-of-the-art Machine Learning and Artificial Intelligence capabilities without having to connect to cloud based services.         The Machine Learning Model Playgrounds is a series of Windows programs built using pure python libraries and code. Each of the programs is a user-friendly demo of Image Classification powered by a specific image classification model of popular Machine Learning Algorithms trained on the ImageNet (1000 object classes ) dataset. Each program provides a user interface where users can select a picture from their Windows system folder while the program process the selected picture and give top-10 possible results of the objects detected with percentage probability per each result.           This repository contains the source code, models and builds of each of the programs in the Model Playgrounds series. It is provided to allow other developers outside our team to adapt, modify or extend the code to produce more programs that may be specific to a social, business, economic or scientific need.         The dependencies used for this project are listed below:     - Python 3.5.2     - Tensorflow 1.4.0     - Keras 2.0.8     - Numpy 1.13.1     - Scipy 0.19.1     - wxPython 4.0.0 Below you will find the details and pictures of each of the programs in the series.           The ResNet Playground is powered by the ResNet50 model trained on the ImageNet dataset. You can find its source codes in the resnet-playground folder of this repository or follow this link. You can also download the Windows Installer for the program in the Release section of this project or follow this link.           This program is a Windows 64-bit software that can be installed on Windows 7 and later versions of the Operating System. It has an installer size of 227mb and install size of 690mb. The program was compiled using PyInstaller 3.3 for Python 3.5 .

SRDenseNet-pytorch - SRDenseNet-pytorch（ICCV_2017）

•    Python

The training data is generated with Matlab Bicubic Interplotation, please refer Code for Data Generation for creating training files. The test imageset is generated with Matlab Bicubic Interplotation, please refer Code for test for creating test imageset.

DenseNet-Cifar10 - Train DenseNet on Cifar-10 based on Keras

•    Python

Train the DenseNet-40-10 on Cifar-10 dataset with data augmentation. The implementation of DenseNet is based on titu1994/DenseNet. I've made some modifications so as to make it consistent with Keras2 interface.

3d-DenseNet - 3D Dense Connected Convolutional Network (3D-DenseNet for action recognition)

•    Python

A number of layers, blocks, growth rate, video normalization and other training params may be changed trough shell or inside the source code. There are also many other implementations, they may be useful also.

river_ice_segmentation - River ice segmentation with deep learning

•    Jupyter

unet, tf_unet and image-segmentation-keras/deeplab_keras contain other implementations of these models that did not work as well as the above. The commands for running each model are provided in a .md file in the corresponding folder. For example, commands for UNet and DenseNet are in image-segmentation-keras/unet.md and densenet/densenet.md. The commands are organized hierarchically into categories of experiments and a table of contents is included for easier navigation.