About me

I am a Master's Student in Computer Vision at the Universitat Autònoma de Barcelona. Besides that, I am working as a Research Assistant at the Document Analysis Group (DAG) of Computer Vision Center. Currently, I am working on the document layout analysis and graph reasoning problems.

Beyond the digital world, I find solace in expressing my creativity through sketches and poems. The strokes of my pencil on paper and the words that flow from my pen are my artistic companions, allowing me to convey emotions and stories in a unique way.

Research Interests

  • design icon

    Sketch Generation

    Working on realistic sketch generation with boundless creativity through diffusion models.

  • Web development icon

    Document Layout Analysis

    The structured insights of documents with seamless comprehension.

  • mobile app icon

    Text Spotting

    Unveils hidden words in a visual feast.

  • camera icon

    Bioinformatics

    Cracking the code of life, where genes tell their tales.

Recent News

  • Daniel lewis

    Master Thesis Dissertation

    Enhancing Document Layout Analysis using Transformer-Based Semi-Supervised Learning and Graph-Based Knowledge Distillation.

  • Jessica miller

    ICDAR 2023

    SwinDocSegmenter: An End-to-End Unified Domain Adaptive Transformer for Document Instance Segmentation (Oral Presentation). For more details please visit: github.com/ayanban011/Swindocsegmenter

  • Emily evans

    Trans. Image Processing

    A New Language-Independent Deep CNN for Scene Text Detection and Style Transfer in Social Media Images.

  • Daniel lewis

    MDPI Genes

    PPIGCF: A Protein–Protein Interaction-Based Gene Correlation Filter for Optimal Gene Selection.

Daniel lewis
quote icon

Daniel lewis

In my groundbreaking master thesis dissertation, I delved into the realm of document layout analysis, a crucial facet of modern information processing. Focusing on innovation and efficiency, I harnessed the power of cutting-edge Transformer-based semi-supervised learning techniques. By integrating Graph-Based Knowledge Distillation into this approach, I not only refined the understanding of document layouts but also paved the way for enhanced analysis accuracy. My research successfully demonstrated the potential of this amalgamation, providing a robust framework that pushes the boundaries of what's achievable in document layout analysis. This breakthrough is a testament to the convergence of AI methodologies, propelling the field forward and promising a future where document analysis is more insightful and precise than ever before.

Resume

Education

  1. Universitat Autònoma de Barcelona

    2022 — 2023

    Master of Science in Computer Vision (CGPA: 7.31/10)

  2. Jalpaiguri Government Engineering College

    2017 — 2021

    Bachelor's of Technology in Computer Science & Engineering (CGPA: 9.15/10)

Experience

  1. Research Assistant @ Computer Vision Center

    June, 2022 — Present

    Analysis of vision and language model for instance document layout segmentation and visual question answering.

  2. Programmer Analyst Trainee @ Cognizant

    Sep, 2021 — Sep, 2022

    Developing a robust pension management system using Spring Boot and Maven in Java.

  3. Project Linked Person @ Indian Statistical Institute, Kolkata, India

    Mar, 2022 - May, 2022

    Analysis of effective deep learning techniques for natural scene text detection, recognition, and handwritten answer script evaluation.

  4. Junior Project Engineer @ Indian Institute of Technology, Kharagpur, India

    Aug, 2021 - Feb, 2022

    Remote monitoring and real-time control of defects in friction stir welding process and preventive health monitoring of friction stir welding machine.

  5. Data Science Intern @ Augray, Chennai, India

    July, 2020 - Aug, 2020

    Automate the 3d annotation process and data collection and preprocessing, exploratory data analysis for augmented reality to obtain virtual trial on shoes.

Awards and Achievements

  1. Swami Vivekananda Merit and Means Scholarship

    2017 — 2021

    A grant from Directorate of Technical Education (DTE) from Government of West Bengal, India.

  2. Bengalathon, Finalist

    2019

    State Level Hackathon, Organised by Department of Information Technology, Government of West Bengal, India.

  3. Intel Edge AI Scholarship,, Winner

    2019

    A financial aid from Udacity to learn OpenVINO toolkit.

  4. Google Build for Digital India, finalist

    2020

    An enriching journey with rich technical resources to learn the latest in tech.

  5. AWS Machine Learning Scholarship, Winner

    2021

    A financial aid to learn foundational ML concepts

  6. Researchathon, Winner in healthcare track

    2021

    An international hackathon organized by United Nations Environment Program and Harvard Kennedy School.

  7. Nexas-A-Thon, 2nd in waste management track

    2021

    A national hackathon organized by organised by Indian Institute of Technology Guwahati.

  8. INDIAai Lab2Market, Winner

    2021

    a program organized by NASSCOM aimed at helping students and faculty in Indian academic space translate their academic research to products.

My skills

  • Programming Languages: C/C++, Java, Python, R, Matlab
    80%
  • Databases: Mysql, sqlite.
    70%
  • Data Science Library: numpy, pandas, matplotlib, seaborn, scikit-learn, opencv, folium, tkinter.
    90%
  • Deep Learning Library: Keras, Tensorflow 2.0, Pytorch.
    90%
  • Web Dev: Html, css, JavaScript, jQuery, bootstrap.
    50%
  • Misc.: Linux: Shell Scripting, git/github.
    60%

Blogs

Contact

Contact Form