Skip to content
View KAVYANSHTYAGI's full-sized avatar
🏠
Working from home
🏠
Working from home

Block or report KAVYANSHTYAGI

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Please don't include any personal information such as legal names or email addresses. Maximum 250 characters, markdown supported. This note will be visible to only you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
KAVYANSHTYAGI/README.md

Hi, I'm Kavyansh Tyagi

AI Researcher • 3D Medical Imaging • Neuromorphic & Graph Learning

3rd Year Undergraduate Researcher at NIT Kurukshetra (India)


🎓 About Me

I am an undergraduate researcher focused on efficient deep learning architectures for vision and scientific AI.
My research spans 3D medical image segmentation, neuromorphic liquid neural networks, and graph-based learning systems, with applications in healthcare and hyperspectral analysis.

Research Highlights:

  • 📚 Publications: Springer & Elsevier journals
  • 🛠️ Open Source: Contributor to PyTorch Lightning & OpenCV & SnnTorch
  • 🔬 Focus: Building architectures that deliver higher segmentation accuracy than existing state-of-the-art approaches with lower computational costs

🛠️ Technologies & Tools


📊 GitHub Statistics


🔬 Featured Research Projects

🧠 RefineFormer3D: Efficient 3D Segmentation

Tech Stack: PyTorch, CUDA, Medical Imaging Libraries
Contribution: A lightweight transformer-based model achieving higher segmentation accuracy than existing state-of-the-art approaches while reducing GFLOPs and parameter count
Impact: Strong results on BraTS & Synapse datasets

📡 Spectral Causal Neural Field (SCNF)

Tech Stack: Torch, Fourier Operators, Spectral Analysis
Contribution: Physics-aware causal framework for hyperspectral super-resolution
Impact: Benchmarking against DiffuHSI and RAWformer

🔬 LiquidSpikeFormer

Tech Stack: PyTorch, SpikingNN, Differential Equations
Contribution: Combines liquid neural networks with transformer dynamics for adaptive temporal modeling
Impact: Neuromorphic computing applied to event-based vision

🌐 Temporal Graph Networks

Tech Stack: PyTorch Geometric Temporal, NetworkX
Achievement: +15% over baselines on large-scale Wikipedia dataset (5M+ nodes)
Status: NeurIPS 2025 target


📈 Contribution Graph


🏆 Achievements & Recognition

  • 📚 Published in Springer & Elsevier journals
  • 🌟 Contributor to PyTorch Lightning & OpenCV & SnnTorch
  • 🔬 Targeting top-tier Core A* conferences (NeurIPS, CVPR, ICLR)
  • 💡 Developed novel architectures in 3D vision, neuromorphic AI, and hyperspectral learning

📫 Connect

Currently Seeking: Summer 2025 AI/ML Research Internships
Interests: Computer Vision • Neuromorphic AI • Graph Learning • Medical Imaging


"Advancing efficient AI architectures from theory to real-world impact"

Popular repositories Loading

  1. LiquidSpikeFormer LiquidSpikeFormer Public template

    Jupyter Notebook 3

  2. GAN-is-all-you-need GAN-is-all-you-need Public

    This repository has all the GAN models built from scratch that are useful in various domains from generating image, video, text or even new sampled tabular data. The main focus of this repo will be…

    Jupyter Notebook 2

  3. SIH2024_AI-INNOVATORS SIH2024_AI-INNOVATORS Public

    Jupyter Notebook 1 1

  4. Emotional-Analysis Emotional-Analysis Public

    This project employs a deep learning model with LSTM and machine learning techniques to classify sentences into 33 emotions. A partially self-created dataset enhances model training, enabling nuanc…

    1

  5. embeddedml embeddedml Public

    Forked from jonnor/embeddedml

    Notes on Machine Learning on edge for embedded/sensor/IoT uses

    Jupyter Notebook 1

  6. Generative-AI-Mastery-Resources Generative-AI-Mastery-Resources Public

    Forked from entbappy/Generative-AI-Mastery-Resources

    Jupyter Notebook 1