jvirico/gans-keras

Stars
3
Rank 3,957,292 (Top 79 %)
Language
Python
License
MIT License
Created over 3 years ago
Updated over 3 years ago

jvirico/gans-keras

jvirico

There are no reviews yet. Be the first to send feedback to the community and the maintainers!

Multilayer Perceptron GAN, and two Convolutional Neural Network GANs for MNIST and CIFAR.

kalman-tracker

An algorithm for object tracking based on Kalman Filter is implemented using OpenCV C++ library. Two modes of operation are coded, a Constant Velocity Model, and an Acceleration Model.

mouse_CT_3D_reconstruction

3D reconstruction of a mouse from a collection of Sinograms coming from a Computed Tomography scan, with Filtered Backprojection and Matlab.

slam-navigation

SLAM navigation on simplified scenario (FastSLAM implementation using Python) based on Particle Filter (Sequential Monte Carlo). What happens when the visual support of a drone is missing?

Jupyter Notebook

normality-tests-pvalues-boxcoxtransformations

Strategies for analyzing the distribution of datasets, switching the data towards a normal distribution testing different manual transformations and Box-Cox transformation.

scania-truck-failure-prediction

Data Mining and Machine Learning APS Failure at Scania Trucks Data Set.

Jupyter Notebook

point-cloud-manipulation

Voxels manipulation and mesh generation using Python.

binary_expression_tree

Java implementation of a Binary Expression Tree to manage algebraic expressions using Composite and Visitor design patterns.

pytorch-templates

Code templates for common Neural Networks in PyTorch, including Perceptron, MLP, Radial Basis Function Nets, GANs...

augmented-reality-poc-unity

AR app built on Android’s ARCore that allows to virtually place multiple true-to-scale 3D object models in a real environment using a mobile phone. By simply uploading new objects to the cloud-based catalog of 3D objects, it can be easily extensible to other use cases for fast creation of virtual prototypes on top of existing real spaces.

composite-pattern

Small examples of Composite design pattern implemented in Java and Python.

graph-conv-neural-nets

sift-feature-matching

SIFT feature matching using OpenCV.

ACGAN_Chromos

Adaptation of ACGAN to fit Chromosome images for Data Augmentation and Chromosome type conditioning.

object_detection_classification

Object detection and classification using Computer Vision, OpenCV, and C++. Implements: (1) blob extraction using SequentialGrass-Fire algorithm, removing the blobs with a size below a certain threshold to eliminate noise; (2) Blob classification using Aspect Ratio feature and simple statistical classifier; (3) Implementation of extraction of stationary foreground pixels based on foreground history; (4) Custom implementation of Grass-Fire algorithm without using OpenCV’s in-built Connected Component Analysis functionalities. (5) Attempt to improve blob classification by using statistical properties of color channels. The implementation is tested on two datasets.