Vítor Godeiro

I am a machine learning engineering at AcertPix in São Paulo, where I work on computer vision and machine learning.

I did my bachelor's degrees in Computer Science from Federal University at Rio Grande do Norte, where I was advised by Profs. Bruno Motta de Carvalho and Bruno Santana da Silva. I did my MSc. in Computer Science at Institute of Informatics in the Federal University at Rio Grande do Sul, where I was advised by prof. Manuel de Oliveira

Email  /  CV  /  Google Scholar  /  Lattes  /  Linkedin

I'm interested in computer vision, machine learning, optimization, image processing, computational photography and human-computer interaction.

We present a real-time technique for simulating accommodation and low-order aberrations (e.g., myopia, hyperopia, and astigmatism) of the human eye. Our approach models the corresponding point spread function, producing realistic depth-dependent simulations. Real-time performance is achieved with the use of a novel light-gathering tree data structure, which allows us to approximate the contributions of over 300 samples per pixel under 6 ms per frame. For comparison, with the same time budget, an optimized ray tracer exploring specialized hardware acceleration traces two samples per pixel. We demonstrate the effectiveness of our approach through a series of qualitative and quantitative experiments on images with depth from real environments. Our results achieved SSIM values ranging from 0.94 to 0.99 and PSNR ranging from 32.4 to 43.0 in objective evaluations, indicating good agreement with the ground truth.

We proposed a method for segmenting and classifying thin rock sections into pores and background (comprised of rocks and cement), by using a procedure that performs color space reduction on the CIELab space and pixel classification using convolutional networks.

We proposed a method for segmenting and classifying chronic ulcers into three types of tissues. Our method consists of a step for reducing the noise, followed by the watershed based segmentation of the ulcer, color space reduction on the CIELab space and tissue classification using the U-Net.

Where Open Stores In São Paulo?
Vítor Godeiro, 2023

The project analyzes data from the city of São Paulo to recommend specific locations for opening a restaurant in a given neighborhood, targeting adults aged 25 to 50 from social classes A (income levels A1 and A2) and B (income levels B1 and B2). Based on its operation in the city of Rio de Janeiro.

Raw Image Decoder
Vítor Godeiro, 2019

This project is a raw image decoder implemented in Python. This code convert the raw image into full color image. In this pipeline we read file in pattern [R G G B] and perform the bilinear demosaic then we can choose the white balance algorithm between Gray World, White Patch, Iterative and Percentile. Finally the code run the gamma correction and give the result image.

Vector Graphics Rendering
Vítor Godeiro, 2018

Vector Graphics Rendering developed in LUA at 2D Computer Graphics summer course at IMPA. This render is able to draw quadratic and cubic Bézier curves, transparency, linear and radial gradients, supersampling and acceleration datastructures.

Driver Drowsiness Detection
Vítor Godeiro, 2017

This is a project implementing Computer Vision techniques to detect drowsiness of a driver and emit sound and luminous alert. This code can detect your eyes, mouth, head tilt and alert when the user is drowsy or sleeping.

Maze
Vítor Godeiro, Leno Pereira, 2015

This is a game of maze. The maze generation is randomizes using disjoint sets, the exit path is found by backtracking. In this game it is possible to play against an artificial intelligence, the difficulty of artificial intelligence is based on the coefficient of intelligence, which influences the time to make a decision, the ability to make the right decision and the remembrance of the already exploited cells.

Flood-IT
Vítor Godeiro, 2015

Flood It is a simple strategy game in which you have to flood the whole game board with one color in less than the permitted steps. The project was developed in C++ using the Simple and Fast Multimedia Library (SFML).

GPU-FPGA
Vítor Godeiro, Samuel Natã, 2015

This project implement the GPU in FPGA Altera DE0 Board. The GPU should use a framebuffer with resolution capability 90x90 pixels. The project draws Lines and Circles.

Erdös = 4

I like this website