Bhalaji Nagarajan

Postdoctoral Researcher at Barcelona Supercomputing Center, collaborating with AIBA

bhalaji.nagarajan@ub.edu
LinkedIn
Twitter
Research Gate
Google Scholar

Profile summary

"Dr. Bhalaji is a Postdocoral Researcher at Barcelona Supercomputing Center (BSC) collaborating and working under the supervision of Prof. Petia Radeva at the University of Barcelona. He finished his PhD Thesis titled "Mastering the Traid of Data, Models and Tasks in Deep Learning for Image Understanding" with "Cum Laude" in June 2024. His academic pursuits revolve around applying machine learning and deep learning to address social issues to enhance the quality of life for individuals. His thesis line is centered on pioneering data-centric deep learning methodologies, with a particular focus on visual food analysis. His research interests span self-supervised learning, learning with noisy labels, and uncertainty estimation.

Bhalaji received the FPI grant from the Ministry of Science and Innovation (MICINN), Spain and the FI grant (Resigned) from the University and Research Grants Management Agency (AGAUR), Catalonia. He previously held the pre-doctoral contract linked with “ICREA ACADEMIA 2014”. Prior to his doctoral studies, he held research positions at Corporate Research, Robert Bosch, India and Dept. of Electrical and Electronics Engineering, Amrita University, India. He holds a Bachelors and Masters degree in Computer Science and Engineering, both from Amrita University, India. Before starting a career in deep learning and machine learning research, he worked in various software roles across different companies."

Latest publications

Bhalaji Nagarajan has 33 publications on Google Scholar.
Only the last 20 publications are shown here. Access Bhalaji Nagarajan Google Scholar Profile

Learning from Semantic Dictionaries: Discriminative Codebook Contrastive Learning for Unified Visual Representation and Generation

2026 | Imanol G Estepa, Jesús M Rodríguez-de-Vera, Bhalaji Nagarajan, Petia Radeva | arXiv preprint arXiv:2605.25012, 2026

Bhalaji Nagarajan

Profile summary

Latest publications

Learning from Semantic Dictionaries: Discriminative Codebook Contrastive Learning for Unified Visual Representation and Generation

Anatomically guided latent diffusion for high-resolution 3D chest CT synthesis

Conjuring Positive Pairs for Efficient Unification of Representation Learning and Image Synthesis

Adaptive vision-language prompt learners for learning with noisy labels

Precision at scale: Domain-specific datasets on-demand

Dual Polarity Prompts with Stochastic Entropy Perturbation for Label Noise

CLIP-DoRA: Weight-decomposed Low-rank Adaptation for Efficient Vision-Language Models

Decoding class dynamics in learning with noisy labels

LOFI: LOng-tailed FIne-grained network for food recognition

Bayesian DivideMix++ for enhanced learning with noisy labels

CCLM: Class-Conditional Label Noise Modelling

All4One: Symbiotic Neighbour Contrastive Learning via Self-Attention and Redundancy Reduction

ELFIS: Expert Learning for Fine-grained Image Recognition Using Subsets

Dining on Details: LLM-Guided Expert Networks for Fine-Grained Food Recognition

Deep ensemble-based hard sample mining for food recognition

F2-Net: Feature Finding Network for Long-Tailed Fine-Grained Food Recognition

Good Fences Make Good Neighbours

Bayesian deep learning for semantic segmentation of food images

Learning Multi-Subset of Classes for Fine-Grained Food Recognition

Uncertainty-aware selecting for an ensemble of deep food recognition models

Research lines

Deep Learning

Uncertainty Modeling