Developed an artificial intelligence approach to accurately detect cassava leaf diseases by enhancing weak signal patterns in agricultural imaging.
Investigated the dynamics of social systems influenced by age groups and law enforcement through mathematical modeling.
Proposed the use of compact natural language models for efficient financial sentiment analysis in resource-constrained environments.
Modeled how law enforcement interventions affect social behavior and optimal response strategies.
Demonstrated how data enhances predictive models for disease spread and control measures.
Introduced a computer vision model for accurate crop disease identification suitable for rural and low-infrastructure regions.
Focused on evaluating and improving visual quality in AI-generated images using efficient vision transformers.
Designed compressed and quantized models for land cover and maritime image classification in Earth observation.
Proposed optimal model pruning and resolution scaling techniques for energy-efficient deployments in remote sensing.
Demonstrated a real-time medical diagnosis tool for cardiac disorders deployable on mobile or wearable devices.
Proposed intelligent quantization techniques for optimizing deep learning model size without losing accuracy.
Introduced a novel motion estimation technique to compress surveillance video data efficiently.
Built compact emotion recognition systems for speech analysis using adaptive layer quantization.
Combined EEG and EOG modalities with machine learning for better sleep disorder classification.
Designed a deep learning approach that uses signal transformation for early seizure detection in clinical settings.
Enhanced sleep stage classification by selecting relevant EEG and EOG features and improving temporal awareness.
Explored the role of cyclic alternating patterns in identifying insomnia-related disruptions.
Introduced a novel LightGBM-based classifier utilizing EOG signals to achieve efficient and interpretable diagnostics for a range of sleep disorders.
Proposed a comprehensive framework that integrates feature selection across EEG and EOG modalities, temporal modeling, and data augmentation to enhance accuracy in automatic sleep stage classification.
Student(s):Shivaanee Eswaran
Work:Uncertainty-guided Style-aware Perceptual Quality Assessment for AI-Generated Images
Student(s):Ahmed Silima Vuai
Work:Efficient Land-Cover Image Classification via Mixed Bit-Precision Quantization
Student(s):Ahmed Silima Vuai and Shivam Bhardwaj
Work:Fully funded research stays for one semester at Grenoble INP, France, and Université de Nantes, Polytech Nantes, France
Student(s):Vikalp, Shivam, Patrick, and Madhusudan
Work:VRD Layout Analysis and Structure Parsing competition
