Skip to content

Research

Peer-reviewed work on deep learning, vision transformers and quantum machine learning, mostly applied to satellite imagery.

Peer-reviewed papers
4
Journal articles
2
Conference chapter
1
Conference paper
1

Publications

2026

Journal

Quantum Convolutional Neural Network for Slum Classification using Sentinel-2 Imagery

Archana G. Pai, Rupesh Kumar Yadav Mediboyina, Krishna Mohan Buddhiraju, Surya S. Durbha
Quantum Information Processing, 25(4), Article 119
DOI ↗Paper ↗

This study explores Multi-channel Quantum Convolutional Neural Networks for slum classification using Sentinel-2 satellite imagery. The work evaluates a hybrid quantum-classical model based on Multi-Channel Quantum Image encoding against a structurally comparable classical CNN, with stratified sampling used to address class imbalance.

Journal

Slum Detection Using U-Net Attention Gate on Mid-Resolution Satellite Imagery

Archana G. Pai, Aaron Augustine, Rupesh Kumar Yadav Mediboyina, Ankit Jha, Krishna Mohan Buddhiraju, Surya S. Durbha
Journal of the Indian Society of Remote Sensing, pp. 1–14
DOI ↗Paper ↗

Slum detection has emerged as a critical area of research in urban studies and remote sensing, driven by the need for effective urban planning, resource allocation, and policy intervention during disaster management. Detecting slums from satellite imagery presents a significant challenge due to variations in building materials, density, spatial organization, and local environmental factors, making it difficult to establish a universal detection approach. This study makes use of all bands of Sentinel-2 up to 20 m along with four spectral indices and an ignore mask to detect slum pockets in Mumbai. An attention mechanism is incorporated into U-Net architectures to enhance feature representation and improve classification performance. The model was further evaluated for generalization and transferability on a patch from the Jakarta dataset, where it demonstrated strong performance in slum classification. This work can be used by government bodies to continuously track and monitor the growth of slum areas and assist decision-makers in making informed decisions. Our proposed method with the attention module focuses on the most relevant spatial and contextual information within the image, strengthening discriminative feature learning and achieving superior results for slum detection.

My role (co-author): validated the methodology, implemented and corrected the codebase, and fixed errors in the model/training code to ensure reproducible results.

Published 18 June 2026.

2025

Chapter

Enhancing Hyperspectral Imagery Classification using CNN and ViT

Rupesh Kumar Yadav Mediboyina, Rishika Ayyalapu, Sanjana Gose, B. N. Shashank, S. Nagesh Bhattu
Pattern Recognition and Machine Intelligence (PReMI), Springer LNCS, pp. 323–332
DOI ↗Paper ↗

This work studies hyperspectral image classification using convolutional and transformer-based architectures, with attention to spectral-spatial preprocessing and low-data classification performance.

Conference

Graph-Based Stance Grouping in Multi-Participant Discussions

Lamim Zakir Pronay, Mohan Koruprolu, Rupesh Kumar Yadav Mediboyina, Venna Nithin Krishna, Sathvik Narala, Vishwachandra Moola
Advances in Computer Science Research, Proceedings of the International Conference on Sustainability Innovation in Computing and Engineering (ICSICE 2024), pp. 1572–1582
DOI ↗Paper ↗PDF

Motivated by the need for identifying and propagating stances along debate threads where this correspondence can be elusive, we present a stance propagation model built on the Heat Diffusion algorithm, and show that it can effectively spread stances over the nodes of a debate thread. Our authors are represented by the nodes of our model, posts written are oriented to the edges in this case, data has been retrieved from the CreateDebate platform. The experimental results show that the proposed model achieves comparable performance as the current state-of-the-art models. This algorithm has shown to be very effective at growing stances in unsupervised settings and, as such, these results provide valuable insight into the evolution of online debates.

Paper collections

Reading lists I keep up to date alongside my own work, for anyone starting out in these areas.