Muhammad Asad Ali

About

Machine Learning Researcher at DFKI, working on 3D Computer Vision in Mixed Reality with Prof. Dr. Didier Stricker. My previous experiences include working with Prof. Dr. Faisal Shafait on various Computer Vision R&D and commercial Projects, particularly in Document Image Understanding domain.

Python
C++
Pytorch
Flask
Javascript
ReactJS
Machine Learning
3D Computer Vision
Digital Image Processing

Work Experience

Dec 2021 - Mar 2024
Graduate Research Assistant
  • - Multi-view 3D Hand Pose Estimation
  • - 3D Human Pose Estimation
Nov 2020 - Aug 2021
Associate Researcher
  • - Medical Document Classification & Segmentation
  • - Structured Data Extraction from Unstructured Documents
Jun 2019 - Oct 2020
Machine Learning Engineer
  • - Table Detection and Structure Extraction
  • - QuickAnnot tool for table data annotation
  • - Pictogram Detection and Classification
  • - TesseractOCR Wrapper to reconginize chemical names of medicines with greek letters
  • - UrduOCR web demo for handwritten and machine-printed urdu text recognition

Education

May 2024 - Present
PhD in Computer Science
  • 3D Computer Vision
    Pose Estimation
Oct 2021 - Mar 2024
Masters in Computer Science
  • Intelligent Systems
    Visualization & Scientific Computing
  • German Grade: 1.4
  • Thesis: Robust 3D Hand Pose and Shape Estimation through Cross-view Fusion
Sep 2015 - Jun 2019
Bachelors in Computer Science
  • Computer Vision
    Network Security
  • German Grade: 1.1
  • Thesis: Automated Gait Recognition

Publications

No publications this year.

Certificates

Projects

Hand Pose Estimation

Estimating robust 3D hand pose and shape in real-time by utilizing multi-view camera inputs through cross-view fusion

3.27 times faster than the state-of-the-art approach at that moment.


Automated Gait Recognition

Gait recognition is a cutting-edge biometric technology that identifies individuals by their unique walking patterns using deep learning.

This project was nominated for the gold medal award for best bachelor's thesis


Table Structure Extraction

Table Detection and Structure Extraction using an end-to-end architecture of Graph Neural Networks (GNNs) and Convolutional Neural Networks (CNNs).


Download CV