Biography

I got my M.S in Artificial Intelligence at Informatics Center of Federal University of Pernambuco (CIn - UFPE) under the supervision of Professor Tsang Ing Ren and Professor Nima Kalantari (Texas A&M University). I graduated from Informatics Center of Federal University of Pernambuco (CIn - UFPE) where I majored in Computer Engineering. At UFPE, I worked with Professor Tsang Ing Ren on Monte Carlo denoising. My research includes Deep Learning applied to Computer Graphics, Computational Photography and Computer Vision. I also work as Software Engineer Consultant at Vektore helping them to improve their data rendering/processing system. You can find more information on my resume.

Interests

  • Computer Vision
  • Computer Graphics
  • Computational photography
  • Deep Learning

Education

  • M.S. Computational Science & Engineering, 2020
    Universidade Federal de Pernambuco
  • B.A. Computer Engineering, 2018
    Universidade Federal de Pernambuco

News
Sep/2020 I started working at Dr. Vladlen Koltun's Intelligent Systems Lab at Intel.
Aug/2020 I successfully defended my M.S. thesis!
Aug/2020 One paper accepted at SIBGRAPI 2020!
July/2020 Starting in September I will be joining Dr. Vladlen Koltun's Intelligent Systems Lab at Intel as Research Scientist resident.
Mar/2020 One paper conditionally accepted to SIGGRAPH 2020!
Nov/2019 I am looking for research internship for Summer 2020!
Aug/2019 I started working in the Aggie Graphics Group at Texas A&M University.
Jun/2019 I will be working for 8 months in the Aggie Graphics Group at Texas A&M University as Assistant Research Scientist under supervision of professor Nima Kalantari.

Publications

Single Image HDR Reconstruction Using a CNN with Masked Features and Perceptual Loss.

SIGGRAPH, 2020

SANTOS, M., TSANG ING, and NIMA KALANTARI

An Analytics Framework for Augmented Reality Applications

Journal on Interactive Systems, 2018

TEIXEIRA, J., FIGUEIREDO, L., MAGGI L., TEICHRIEB, V., SANTOS, M., AND ARAUJO, C.

Analyzing AR viewing experience through analytics heat maps for augmented content

19th Symposium on Virtual and Augmented Reality, 2017

SANTOS, M., TEIXEIRA, J., FIGUEIREDO, L., TEICHRIEB, V., AND ARAUJO, C.

Projects and Experiences

Experiences

Assistant Research Scientist (Intern)
Texas A&M University
Aug 2019 - Mar 2020, College Station/TX

Deep Learning applied to Computer Graphics and Computational Photography. Hosted by Professor Nima Kalantari.

Student Researcher
OKI Brasil
Jun 2017 - Oct 2018, Recife/PE

Development of a biometric system using machine learning and computer vision techniques with Keras, OpenCV, and scikit-learn libraries.

Student Researcher
Simplifique GP
2016 - 2017, Voxar Labs (Recife/PE)

Developed a data intensive rendering system to iPad device using C++ and Objective-C.

Student Researcher
LG Eletronics
2014 - 2015, Voxar Labs (Recife/PE)

Developed a image enhancement system to Android using Computer Vision and Image Processing techniques, OpenCV, C++, and Android NDK and JNI.

Major Projects

Path Tracer Denoiser (B.S final project)

Path tracing can deliver beautiful images. However, it needs thousands of samples per pixel to generate good results. Using low samples per pixel results in noisy images and the traditional denoising techniques are either limited or need a very trick parameters tuning. This project consists on a CNN which delivers a filter able to generate noise-free images from noisy ones.

Deep Shading: Convolutional Neural Networks for Screen-Space Shading

Based on Nalbach et al. 2017 paper. In this project, a set of buffers are provided to a CNN to synthetize differend shading effects (such as Ambient Occlusion, Depth of Field, Global Illumination and Sub-surface Scattering).

Human Skin Detection Using Bayesian Approach

We have prior histograms to classify skin and not skin pixels. The histograms were built from a YCbCr dataset of manually classified pixels. The algorithm is able to detect human skin even with changes in environmental light conditions. Our algorithm works well in many different scenarios and is fast.

Contact Me

Center of Informatics - CIn/UFPE
Av. Jornalista Anibal Fernandes, Cidade Universitária
ZIP Code: 50740-560
Recife/PE - Brazil
E-mail: mss8 [at] cin [dot] ufpe [dot] br