Research
Publications
Peer-reviewed research in computer vision and domain adaptation for deep learning systems.
Improved Discrepancy based Domain Adaptation Network using Combined Loss Functions and Feature Transformations
ICVGIP '23: Proceedings of the Fourteenth Indian Conference on Computer Vision, Graphics and Image Processing
This paper presents an improved domain adaptation network that leverages combined loss functions and novel feature transformations to significantly reduce the discrepancy between source and target domain distributions. The proposed approach demonstrates superior performance on standard computer vision benchmarks, achieving state-of-the-art results in cross-domain object recognition tasks.
Improved Domain Adaptation: A Survey
Conference Proceedings
A comprehensive survey of improved domain adaptation techniques in deep learning, covering methodological advances in discrepancy-based methods, adversarial adaptation, self-supervised approaches, and their applications in real-world computer vision systems. The survey identifies open research challenges and future directions in the field.