Publication

Active-BrainNet : An Efficient Active Learning Paradigm For Brain Tumor Classification

A novel Human-in-the-Loop active learning framework for brain tumor classification that requires minimal labeled data while incorporating human expertise. The method introduces three key innovations: an Active Uncertainty Scoring strategy combining informativeness and representativeness for optimal sample selection, an OpAct Thresholding mechanism to determine ideal labeling quantities, and an Active Self Breaking Loop for cycle optimization.

  • Python
  • OpenCV
  • AWS SageMaker
Visit Site → GitHub
Project 1

Professional Experience

Creative Manager — Camber Racing

March 2022 - May 2024

Managed the visual brand and creative direction for an automotive racing team. This role involved bridging technical automotive concepts with engaging visual storytelling.

  • Brand Identity & Creative Strategy
  • Video Production & Post-Production
  • Team Media Leadership
Camber Racing Work

Core Technical Skills

Programming: Python, C, C++, MATLAB, SQL, JavaScript

Deep Learning: TensorFlow, PyTorch, Keras

Vision & 3D: OpenCV, MediaPipe, Jetson Nano

DevOps & Tools: AWS SageMaker, Prometheus, Grafana, Docker

Development Tools: Git, Jupyter, Google Colab, Linux, VS Code

Additional Tools: Microsoft Office (PPT, Excel, Word), Adobe Premiere Pro, Adobe Photoshop, Davinci Resolve

Languages Known

Malayalam (Native)
English (C1)
Hindi (B2)
Tamil (A2)
German (A2)

My Projects

Gesture Control Karate Gameplay System

This project uses computer vision and machine learning to control a karate game using real-time body gestures. It captures your pose via a webcam, classifies the move (e.g., Punch, Kick, Crouch), and simulates the corresponding keyboard press.

  • Python
  • OpenCV
  • Mediapipe
  • PyDirectInput
Visit Site → GitHub
Project 1

Real-Time Animal Intrusion Detection

A real-time wildlife surveillance system built using YOLOv5 and deployed on the NVIDIA Jetson Nano. Trained on a custom dataset of 15+ animal classes—including tigers, lions, giraffes, and monkeys—the model achieved 90% accuracy in detecting and classifying animal activity. Ideal for preventing human-wildlife conflicts through automated monitoring.

  • Python
  • YOLOv5
  • OpenCV
  • NVIDIA Jetson Nano
Visit Site → GitHub

Editing using MoviePy

This project demonstrates automated video editing capabilities using Python's MoviePy module, a powerful library designed to simplify complex video editing tasks. The application can programmatically cut and rearrange video clips, add sound and visual effects, and perform audio editing operations, making it ideal for batch processing multiple video files efficiently.

  • Python
  • MoviePy
Visit Site → GitHub

Instagram Clone Mockup

Created an Instagram UI clone during a 7-day free bootcamp by DevTown. Gained foundational skills in HTML and CSS to build a fully functional, live project on CodeSandbox.

  • HTML
  • CSS
  • CodeSandbox
Visit Site → GitHub

Image Enhancement using MIRNet

MIRNet (Multiscale Information Residual Network) is a state-of-the-art deep learning architecture designed to restore and enhance images. By capturing information across multiple scales, it recovers high-quality content from degraded versions, specifically addressing challenges like low light, noise, and blurring.

  • Python
  • TensorFlow
  • OpenCV
  • Matplotlib
Visit Site → GitHub
Project 3

My Extra-Curricular Activities

Video Editor — Vloggio YT (YouTube)

What started as a creative outlet during the COVID-19 pandemic has evolved into a long-term passion for visual storytelling. For over four years, I have managed the end-to-end post-production for our family YouTube channel, Vloggio YT, transforming raw footage into engaging digital content.

  • Adobe Premiere Pro
  • Adobe Photoshop
  • Canva
  • Adobe After Effects
  • DaVinci Resolve
Visit Site → YouTube

Filmmaker — Pro.Fil (Film Group in Universität Passau)

Collaborating within a dynamic student production team to bring cinematic visions to life through a blend of technical editing and creative direction. Presently completed two short films,'STENOPAIC' (Winter Sem. 2025-2026) and 'Brushed Aside' (Summer Sem. 2025), with a third project in development

  • Adobe Premiere Pro
  • Davinci Resolve
  • Adobe After Effects
Visit Site → GitHub
Project 3