thales-mro/structure-from-motion

In this project, we were given the task of reconstructing the 3D shape of objects in the real world from series of images. To fulfill the requirements, the implemented algorithm deals with mapping of keypoints within a frame, usage of Kanade-Lucas-Tomasi Tracking method for locating those keypoints in other frames and Structure From Motion (SfM) algorithm to map those keypoints to real world coordinates. Also the camera positions were supposed to be recovered. With the coordinates, a 3-D plotter was used to see the shape of the object. Even though the KLT implementation showed to be robust, the SfM algorithm was not correct enough to produce correct 3D shapes of the objects.

Language: TeX - Size: 55.8 MB - Last synced: about 1 year ago - Pushed: about 3 years ago - Stars: 0 - Forks: 0

thales-mro/augmented-reality

In this project, we were given the task of performing Augmented Reality (AR) task in media, in this case, a video. The goal was to make an image appear in every single frame of the video, in this case, a specific target on the video is always substituted by the image, while the camera changes position, orientation and focal distance. In order to fulfill the requirements, the algorithm implemented deals with image descriptors for every frame pair to find the target, a matching algorithm to hypothesize matches and a RANSAC with least squares methodology to calculate the transform between frames. We were able to generate AR video for testing purposes, even though our SIFT implementation was inadequate, and the limitations of our implementation are explained in this report.

Language: TeX - Size: 28.1 MB - Last synced: about 1 year ago - Pushed: about 3 years ago - Stars: 0 - Forks: 0

thales-mro/robot-controller

In this project, the group was given the task of continuing the work done in odometry-and-feature-extraction, by adding basic behaviors to control the P3-DX robot. Three control behaviors are proposed (Avoid Obstacle, Wall Follow and Go To Goal), using models based on Proportional, Integral and Derivative Controller (PID) and Fuzzy logic. Also a behavior coordination strategy was implemented, using state machines. The V-REP simulator was used to validate the developed structures. The group was able to implement the isolated behaviors with good results in their execution cases. For the behavior coordination system, it could handle well cases using the ground truth information, but it had troubles when using the calculated odometry with gyroscope. Further progress synchronizing the sensor reads can be done to improve results.

Language: Python - Size: 7.66 MB - Last synced: about 1 year ago - Pushed: about 3 years ago - Stars: 0 - Forks: 0

thales-mro/odometry-and-feature-extraction

In this project, the group was given task of building an odometry and feature extraction system for an autonomous wheeled robot, P3-DX. The mentioned goals are the basis for robot performance in the environment, as they guarantee robust and reactive operation in real time. Reliable enough odometry leads to trustworthy features, hence more complex actions such as control, mapping and navigation. To fulfill the requirements, the algorithm implemented deals with the kinematic model of the robot for odometry calculation, its sensors for acquiring information from the environment and a RANSAC algorithm to extract features. The V-REP simulator was used to test the developed structure. The group was able to generate different odometry qualities depending on sensor usage, with good results in the best case. The point extraction performs well, considering that the current position of the robot is reliable. The feature extraction algorithm is able to calculate obstacles such as walls and supports of objects, not considering possible outliers

Language: TeX - Size: 3.94 MB - Last synced: about 1 year ago - Pushed: about 3 years ago - Stars: 0 - Forks: 0

thales-mro/video-inpainting

In this project, we were given the task of removing people from a series of videos, with the capability of correctly painting the background behind them properly. To do so, the implemented algorithm uses a pre-trained neural network to detect people within image frames, homography matrix to align those frames, and image ROI to paint the bounding box containing the human beings. A variety of tests were made with indoor and outdoor videos, and the proposed algorithm was able to remove people correctly when using a static camera, however, when the camera is in motion, the inpaiting showed some visual inconsistency.

Language: TeX - Size: 11.7 KB - Last synced: about 1 year ago - Pushed: about 3 years ago - Stars: 0 - Forks: 0

thales-mro/ascii-art

In this project, we were given the task of creating the ASCII art version of images in grayscale and color. To fulfill the requirements, the algorithm implemented deals with filters to resize the input image and to define the colors of the printed characters. The filters built in this work use our implementation of convolution operation when needed. We were able to generate ASCII images in grayscale and color

Language: TeX - Size: 34.2 KB - Last synced: about 1 year ago - Pushed: about 3 years ago - Stars: 0 - Forks: 0

thales-mro/opencv-simple-manip

Language: TeX - Size: 1 MB - Last synced: about 1 year ago - Pushed: about 3 years ago - Stars: 0 - Forks: 0

thales-mro/image-compression

In this project, it was given the task of implementing an image compression algorithm, using mathematical formula- tions. To fulfill the requirements, the solution uses the Principal Component Analysis (PCA) approach. It was able to generate the compressed images for some cases, with some degree of information loss. The details of the implementation are explained in this report.

Language: TeX - Size: 33.4 MB - Last synced: about 1 year ago - Pushed: about 3 years ago - Stars: 0 - Forks: 0

thales-mro/image-measures

In this project, it was given the task of calculating measures of objects within digital images. To fulfill the require- ments, the solution implements four different algorithms, one for each required task from the project description[1]. It was able to generate the outputs needed, and the implementation decisions are explained in this report.

Language: TeX - Size: 8.79 MB - Last synced: about 1 year ago - Pushed: about 3 years ago - Stars: 0 - Forks: 0

thales-mro/image-thresholding

In this project, it was given the task of applying global and local threshold methods to grayscale images. To fulfill the requirements, the solution implements 8 different algorithms, using approaches listed in the literature. It was able to generate the binarized thresholded images, and the differences of each approach, the pros and cons of the implementation are explained in this report.

Language: TeX - Size: 21.5 MB - Last synced: about 1 year ago - Pushed: about 3 years ago - Stars: 0 - Forks: 0

thales-mro/halftoning

Image Halftoning using different techniques based on the literature such as Burke, Floyd, Judice & Ninke, and so on.

Language: TeX - Size: 9.22 MB - Last synced: about 1 year ago - Pushed: about 3 years ago - Stars: 0 - Forks: 0

thales-mro/basic-image-processing

Language: Python - Size: 7.81 KB - Last synced: about 1 year ago - Pushed: about 3 years ago - Stars: 0 - Forks: 0

thales-mro/pyqt-resource-management

Language: Python - Size: 2.43 MB - Last synced: about 1 year ago - Pushed: over 3 years ago - Stars: 1 - Forks: 0

thales-mro/web-dev-learning

Language: HTML - Size: 13 MB - Last synced: about 1 year ago - Pushed: over 3 years ago - Stars: 0 - Forks: 0

thales-mro/test

Language: C - Size: 1000 Bytes - Last synced: about 1 year ago - Pushed: over 3 years ago - Stars: 0 - Forks: 0

thales-mro/non-convex-optimization

Language: Jupyter Notebook - Size: 851 KB - Last synced: about 1 year ago - Pushed: over 1 year ago - Stars: 0 - Forks: 0

thales-mro/odom-p3dx

This repository is used for Autonomous Robotics Course Project 1

Size: 602 KB - Last synced: about 1 year ago - Pushed: over 5 years ago - Stars: 0 - Forks: 0

thales-mro/python-cookbook

Language: Python - Size: 36.1 KB - Last synced: about 1 year ago - Pushed: over 4 years ago - Stars: 0 - Forks: 0

thales-mro/ruby-basic

Repository containing basic Ruby code (4learning)

Language: Ruby - Size: 20.5 KB - Last synced: about 1 year ago - Pushed: over 4 years ago - Stars: 0 - Forks: 0

thales-mro/python-basic

Basic python code to practice syntax

Language: Python - Size: 21.5 KB - Last synced: about 1 year ago - Pushed: over 4 years ago - Stars: 0 - Forks: 0

thales-mro/go-prototypes

Language: Go - Size: 0 Bytes - Last synced: about 1 year ago - Pushed: almost 5 years ago - Stars: 0 - Forks: 0

thales-mro/data-structures

Data structures implemented in C

Language: C - Size: 2 MB - Last synced: about 1 year ago - Pushed: almost 5 years ago - Stars: 0 - Forks: 0

thales-mro/c-basics

It contains basic C code, typically from undergrad courses

Language: C - Size: 543 KB - Last synced: about 1 year ago - Pushed: almost 5 years ago - Stars: 0 - Forks: 0

thales-mro/python_to_c

Simple repository containing algorithms intended for Python programmers migrating to C, for undergrad course

Language: C - Size: 12.7 KB - Last synced: about 1 year ago - Pushed: almost 5 years ago - Stars: 0 - Forks: 0

thales-mro/senior-design

Language: C - Size: 15.3 MB - Last synced: about 1 year ago - Pushed: almost 6 years ago - Stars: 0 - Forks: 0

thales-mro/sparkwebapp

Code example for the Spark web app, including dockerfile

Size: 2.01 MB - Last synced: about 1 year ago - Pushed: over 6 years ago - Stars: 0 - Forks: 0

thales-mro/robotino Fork of ct2034/robotino

Software to run Festo Didaktik Robotino using ROS kinetic

Language: HTML - Size: 60.2 MB - Last synced: about 1 year ago - Pushed: almost 7 years ago - Stars: 0 - Forks: 0