Attention-guided convolutional autoencoder for one-class anomaly detection and localization on CIFAR-10, using CBAM and reconstruction-based scoring.

Improved Multi-modal Image Fusion with Attention and Dense Networks

ROI-guided differential privacy for federated learning on OASIS MRI (Alzheimer’s disease classification) with CBAM attention (Flower + PyTorch)

First-author MDPI Diagnostics (2025) paper on colorectal polyp segmentation using μ-Net with CBAM attention and Explainable AI.

Multi-stage U-Net with CBAM for real-world image denoising

This repository implements a neural network that upscales low-resolution brightness temperature data from AMSR2 satellite observations by 2×, 4×, or 8×. The model uses spatial attention mechanisms and residual learning to reconstruct high-resolution thermal fields while preserving physical consistency. Achieves 40-41 dB PSNR on 2× upsampling.

Dual-stream deep learning architecture for deepfake detection combining spatial (RGB) and frequency-domain (DCT) analysis with CBAM attention mechanisms.

[MICCAI'24] Official implementation of "BGF-YOLO: Enhanced YOLOv8 with Multiscale Attentional Feature Fusion for Brain Tumor Detection".

This framework is a novel Ground Penetrating Radar image analysis framework developed by NS Lab @ CUK, integrating a denoising auto-encoder and channel-wise attention into the YOLOv12 backbone to achieve robust and accurate underground barrier detection under noisy and complex subsurface conditions.

RAG assistant for EU CBAM compliance, built with Snowflake Cortex and Streamlit. It retrieves and interprets indexed policy documents, enabling natural language queries on CBAM rules and emissions guidance. Integrates live EU ETS carbon pricing to support cost estimation workflows.

EfficientNet vs. CBAM: Benchmarking Attention for Ocular Disease Classification

PHP package to read metadata and extract covers from eBooks, comics and audiobooks.

Building a lightweight CNN using the ShuffleNet V2 architecture to estimate object sizes.

Predicting urban heat island for the city of Pune using MODIS LST, PM2.5 and GLC_FCS30D Landcover dataset. The neural network consists of three covolution streams and attention unet for feature fusion and UHI map creation.

Predicting urban heat island for the city of Pune using MODIS LST, PM2.5 and GLC_FCS30D Landcover dataset. The neural network consists of three covolution streams and attention unet for feature fusion and UHI map creation.

2D 수직 촬영 영상(Overhead Imagery)에 특화된 디퓨전 기반 고속·정밀 객체 검출 모델 제안

A comprehensive implementation of CBAM-STN-TPS-YOLO architecture for agricultural object detection, featuring convolutional block attention modules (CBAM), spatial transformer networks (STN), and thin plate spline (TPS) transformations. Includes cross-dataset evaluation on PGP, GlobalWheat, and MelonFlower datasets with statistical validation.

Code for paper "Channel Pruning Guided by Spatial and Channel Attention for DNNs in Intelligent Edge Computing"

Local Intelligence Bootcamp organized by IIT Tirupati in collaboration with Stanford’s Computer Science & Social Good and Women in Computer Science.