Xin Liu

I am a PhD and applied scientist in Computer Vision, Deep Learning, and Computer Graphics. I have been working in these fields since 2004 when I was a student with 8 years of experiences as an applied scientist in industry. I am familiar with c++, python, Caffe, Pytorch, TensorFlow, OpenCV, OpenGL, etc. I am looking for my next job in these fields in Greater Toronto Area.

Staff Machine Learning Scientist 2021 - Present

Compass Management Holdings LLC

Lead R&D efforts in Computer Vision, Deep Learning, and NLP. Representative projects are:

Hero image selection (team lead)
    -     Yolo based object detection, image rating algorithm, and architecture design

Document processing automation (team lead)
    -     BERT model based document classification, Named Entity Recognition.

Senior Machine Learning Scientist II 2019 - 2021

Compass LLC

Lead R&D efforts in Computer Vision, Deep Learning, and Video Generation. Representative projects are:

Listing photo super-resolution (team lead)
   -     SR-GAN based algorithm     
   -     Pytorch, python, AWS SageMaker

Video Studio (team lead) 
   -     Video rendering algorithms based on photos and 3D-models & various AI algorithms
   -     c++, OpenGl, GLSL, python, Pytorch, and AWS

Senior Computer Vision and AI Engineer 2019 - 2019

Aibee US Corporation

Conduct research in Computer Vision and Deep Learning technologies. Representative project:

Object detection, tracking, and pedestrian flow analyses
    -     SSD, Motion Detection, Kalman Filter
    -     Tensorflow, python, and c++

Staff Computer Vision and AI Engineer 2018 - 2019

Huawei Seattle Research Center

Lead R&D efforts in Computer Vision and Deep Learning for cellphone applications. Representative project is:

Real-time AI algorithms on cell phones (team lead)
    -     Super resolution, Denoise, Artifact Removal, Dehaze Networks
    -     Caffe, Tensorflow, and python for deep learning
    -     C++, java, OpenCV, and OpenCL for Android development

Analyst III - Computer Vision and Machine Learning 2014 - 2018

R&D, MDA Corporation

Conduct research in Computer Vision, Deep Learning, Image and Video Processing. Representative projects are:

Ship and wake detection from satellite images (team lead)
    -     Faster R-CNN, and other deep learning based object detection
    -     c++, python, Caffe and OpenCV

Change detection and classification from satellite images
    -     Adaptive image differentiation and deep learning based classification
    -     c++, python, Tensorflow and OpenCV

Resolution enhancement of satellite images (team lead)
    -     Multi-spectrum fusion and image super-resolution
    -     c++, Matlab, Caffe and OpenCV

Moving object detection from airborne videos
    -     Frame registration and differentiation, motion history image, blob detection, classification and object tracking
    -     c++, Matlab and OpenCV

Research Associate 2007 - 2014

University of Calgary

Conduct research in Computer Vision and Computer Graphics. Representative projects are:

Lens effect simulation from synthetic and cell-phone images
    -     Sparse-view based depth-of-field synthesis, depth recovery and parallel algorithms
    -     c++, java, CUDA, OpenCV, and Android SDK/NDK

Real-time Ray Tracing on GPUs
    -     Acceleration structures and algorithms for quick ray-object intersections
    -     c++ and OpenCL for GPU programming

3D reconstruction from 2D images
    -     Shape from X, Delaunay tessellation and refinement
    -     c++, OpenCV, and OpenGL?  NURBS model fusion
    -     Approximation of high-order 3D parametric surface and curves
    -     c++ and OpenGL

PhD in Computer Science

University of Calgary

GDP 4.0/4.0

Focused on Computer Vision, Machine Learning, and Computer Graphics

Master of Science in Computer Science

University of Calgary 2007 - 2009

GDP: 4.0/4.0

Focused on Computer Vision, Machine Learning, and Computational Geometry

University of Calgary, Alberta, Canada 2009 - 2014