I am Bharath Bhushan Damodaran, a Senior Researcher in Machine Learning at InterDigital, Rennes, France. Previously, I was a Post Doctoral Researcher in OBELIX team of IRISA under Prof. Nicolas Courty and Prof. Sebastien Lefevre. I was awarded the Prestige and Marie Curie Post Doctoral Fellowship. I completed my PhD in Remote sensing image analysis under Dr. Ramo Rao Nidamanuri at Indian Institute of Space Science and Technology, Trivandrum. My research interests are applications of machine learning in computer vision and remote sensing image analysis.
Research Interests: Optimal Transport, Deep Learning, Domain Adaptation, Transfer Learning, Semi-supervised Learning, Learning with noisy labels, High dimensional data analysis, Semantic segmentation, Object detection, Hyperspectral Image Analysis, Ensemble learning, Kernel Approximation,
Latest News !!!
11/2019 : Invited to present a poster about my recent research at Data Science for Better Life Workshop, Bayer
11/2019 : Our Paper " Learning with noisy labels" using Entropic OT loss, got accepted at Journal of Computer vision and Image understanding
10/2019 : Presented the research activities of OBELIX team to Indian delegates
02/2019 : Visiting researcher at Prof. Devis Tuia lab, Wageningen University and Research for two weeks
02/2019 : We received positive feed back of our paper learning with noisy labels
01/2019 : Our paper "GEOBIA at the Terapixel Scale: Toward Efficient Mapping of Small Woody Features from Heterogeneous VHR Scenes ", got accepted
08/2018 : Our paper "Unsupervised Domain Adaptation with DeepJDOT" got accepted as oral in NCCV 2018
07/2018 : Our paper "DeepJDOT" on domain adaptation got accepted in ECCV 2018 [codes].
04/2018: I was involved in teaching "EuroSDR-EduServ16" on Topographic Mapping.
03/2018 : Our paper "Terapixel Scale" got accepted in GEOBIA 2018
01/2018: One paper got accepted in ICLR 2018. "Large-scale optimal transport and mapping estimation" openreview link
08/2017: Our paper "Scalable Extraction of Small Woody Features (SWF) at the Pan-European Scale using Open Source Solutions", got accepted in Earth Observation Open Science 2017 conference.
06/2017: joined as Post Doc under Nicolas Courty
03/2017: two IGRASS paper got accepted
01/2017: IEEE TGRS paper "Sparse Hilbert Schmidt independence criterion and surrogate-kernel-based feature selection for hyperspectral image classification " got accepted
12/2016: PERS paper "Attribute profiles on derived features for urban land cover classification " got accepted