(ECCV2020) Tensorflow implementation of A Generic Visualization Approach for Convolutional Neural Networks

(WACV 2020) Tensorflow implementation of Boosting Standard Classification Architectures Through a Ranking Regularizer

No description provided.

"Deep Metric Learning Meets DeepClustering: An Novel UnsupervisedApproach for Feature Embedding" (BMVC 2020)

The t-SNE visualization and actual query results of the deep feature embeddings for the paper "Supervised Deep Feature Embedding with Hand Crafted Feature" that has been accepted by the IEEE Transactions on Image Processing.

A framework for learning low dimensional representations of single-cell images, here applied in enrichment-free liquid biopsies

SVMax : A Feature Embedding Regularizer

Official Implementation of Link Prediction via Node Feature Initialization with LLMs (COLING: Industry Track 2025)

FSE+Attention model for particle identification in ALICE (Pb-Pb Run 3) at CERN. State-of-the-art detector masking with 92.8% accuracy using JAX/Flax. Handles missing detector data through Feature Set Embedding + Multi-head Attention. Production-ready with Focal Loss, class weighting, and two-tier model persistence.

Production-style intent recognition system: transformer embeddings, multi-model inference (LR/SVC), offline STT, dynamic skill modules, and a full Streamlit demo UI. Fast, modular, and extensible. IN SHORT : Modular intent recognition engine with transformers, ML models, and real-time inference.

A deep learning project for similarity measurement and retrieval tasks using feature embeddings.

High level audio features for Javascript

This project builds a deep learning model to detect fake & real Indian currency notes using image data and note-specific features. It includes two training pipelines a base CNN with MobileNetV2 and a two-branch network combining image and feature embeddings. Achieved 99% validation accuracy, proving reliable for real-world currency authentication.