Skip to content
View gadhalekshmip's full-sized avatar

Block or report gadhalekshmip

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.

Maximum 250 characters. Please don't include any personal information such as legal names or email addresses. 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
gadhalekshmip/README.md

Hi πŸ‘‹, I'm Gadha Lekshmi P.

Vision and Image Generation Engineer

πŸ“« Reach Me At:

πŸ“§ gadhalekshmip@gmail.com
πŸ”— LinkedIn
πŸŽ“ Google Scholar
πŸ”¬ ORCID

🎨 Featured AI Art & Generative Projects

Broken Shores | Christmas Special
End-to-end cinematic AI generation pipeline combining custom script-to-storyboard models with diffusion synthesis (Midjourney) and text-to-video generation (Google Veo). Achieved 90%+ frame-to-frame consistency across 200+ frames.

πŸ› οΈ Technical Stack

Generative AI & Creative Systems
GANs (StyleGAN2, Pix2Pix, CycleGAN) β€’ Diffusion Models (Stable Diffusion, DDPM) β€’ Creative Adversarial Networks β€’ Text-to-Image Synthesis β€’ Multimodal Generation Pipelines β€’ Style Transfer & Aesthetic Modeling

Computer Vision
Object Detection (YOLO, Faster R-CNN) β€’ Semantic Segmentation (U-Net, Mask R-CNN) β€’ Keypoint Detection β€’ Vision-Language Models (CLIP, BLIP, LLaVA) β€’ 3D Reconstruction

ML/DL Frameworks
PyTorch β€’ TensorFlow β€’ Hugging Face β€’ OpenCV β€’ MONAI β€’ LangChain

Production & Deployment
AWS (EC2, S3) β€’ Docker β€’ MLflow β€’ BentoML β€’ Streamlit β€’ FastAPI β€’ Terraform β€’ CI/CD (GitHub Actions)


πŸ“š Selected Publications

Journals (Under Review)

  • Landmark-Guided Geometric Warping and GAN Refinement for Controllable and Explainable Cephalometric Profile Transformation
    Electronic Letters on Computer Vision and Image Analysis (ELCVIA), 2025 (Under Review)

  • FedDPI-SH: Quality and Similarity Aware Federated Learning for Medical Image Analysis
    IEEE Journal of Biomedical and Health Informatics, 2025 (Under Review)

  • Multi-Layer Explainability Framework for Semi-Supervised Anomaly Detection in Digital Pathology
    Electronic Letters on Computer Vision and Image Analysis, 2025 (Under Review)

Journals (Accepted)

  • Semi-supervised Teeth Segmentation and Rigid Registration of CBCT–IOS Pairs Using PointNetLK
    MICCAI 2025, Springer LNCS (Accepted) | πŸ† First Place - MICCAI STSR Challenge
    Read here

Conferences

  • From Chaos to Clarity: Satellite Imagery & Deep Learning for Emergency Response
    International Conference on Advances in Aerospace and Energy Systems (IAES 2024) @ ISRO
    Read here

  • Sign2Text: Vision Transformer-Based Sign Language Translation
    IEEE International Conference on Artificial Intelligence in Engineering and Technology (IICAIET 2024)
    Read here


Blog

Pinned Loading

  1. Alien-HR Alien-HR Public

    Python

  2. Sign2Text-Vision-Transformer-Powered-Video-Processing- Sign2Text-Vision-Transformer-Powered-Video-Processing- Public

    Forked from Rohit909-creator/Sign2Text-Vision-Transformer-Powered-Video-Processing-

    Sign Language requires a video modality data to be processed for better accuracy, hence we propose Vision Transformer based Video Processing method. Where each frame of a video is treated as a patc…

    Python

  3. Hierarchical Hierarchical Public

  4. STS_MCCAI_TASK02 STS_MCCAI_TASK02 Public

    Forked from Ajogeorge29/STS_MCCAI_TASK02

    Jupyter Notebook

  5. StyleX-Med StyleX-Med Public

    Hybrid Geometric-GAN Framework for Cephalometric Profile Transformation