Faculty Publications
Permanent URI for this communityhttps://idr.nitk.ac.in/handle/123456789/18736
Publications by NITK Faculty
Browse
2 results
Search Results
Item Deep Learning Techniques for Artistic Image Transformations: A Survey(Institute of Electrical and Electronics Engineers Inc., 2021) Aralikatti, R.C.; Sangeeth, S.V.; Chandavarkar, B.R.Deep learning has greatly revolutionized the ways in which computers tackle problems in vision, speech recognition, machine translation, etc., and has produced results which are almost inconceivable to conventional algorithms. Creative tasks such as fine arts and music composition, which were initially thought to be impossible to computers, are now possible. In this paper, we look at a particular class of problems called image-to-image translation problems and see how it can be leveraged to perform artistic image transformations. Generative Adversarial Networks (GANs) and related neural networks are particularly useful for this task. We explore some of the artistic image transformation tasks that deep learning can be used for and discuss the different machine learning architectures used, the results produced and the advancements made in literature towards tackling such tasks. © 2021 IEEE.Item A Dual-Stage Semi-Supervised Pre-Training Approach for Medical Image Segmentation(Institute of Electrical and Electronics Engineers Inc., 2024) Aralikatti, R.C.; Pawan, S.J.; Rajan, J.Deep neural networks have played a vital role in developing automated methods for addressing medical image segmentation. However, their reliance on labeled data impedes the practicability. Semi-Supervised learning is gaining attention for its intrinsic ability to extract valuable information from labeled and unlabeled data with improved performance. Recently, consistency regularization methods have gained interest due to their efficient learning procedures. They are, however, confined to data or network-level perturbations, negating the benefit of having both forms in a single framework. In light of this, we ask an intriguing but unexplored question: Can we have both network-level and data-level perturbation in the semi-supervised framework? To this end, we present a holistic approach that integrates data-level perturbation in the model pre-training stage, followed by implicit network-level perturbation in the fine-tuning stage. Furthermore, we incorporate networks with manifold learning paradigms throughout the training to facilitate the formation of robust data representations by ensuring local and global semantic affinities adhering to the theory of consensus. Notably, this may be the first attempt in the semi-supervised medical image segmentation archetype to use data and network-level perturbation with a model pre-training strategy. We extensively validated the efficacy of the proposed framework on three benchmark datasets, namely the Automated Cardiac Diagnosis Challenge, ISIC-2018, and Left Atrial Segmentation Challenge datasets, subjected to severely low-sampled labeled data. Notably, in ACDC (4%), ISIC-2018 (5%), and LA (6%) labeled cases, the proposed method outperforms the second-best method by 2.95%, 1.31%, and 0.71% in the Dice Similarity Metric. © 2023 IEEE.