Image Translation

A Generative Adversarial Network or GANs are generative models that use two powerful deep neural networks via an adversarial training regime to draw new samples from an unknown distribution. Apparently, the objective can be portrayed as a tractable approach by minimizing a divergence between the model distribution and real data distribution

Many problems in image processing, computer graphics, and computer vision can be posed as “translating” an input image into a corresponding output image. Automatic image-to-image translation is the task of translating one possible representation of a scene into another, given sufficient training data. Traditionally, each of these tasks has been tackled with separate, special-purpose machinery, despite the fact that the setting is always the same: predict pixels from pixels. However, the model performance deteriorates if data representations are not proper. Our goal is to develop an improved framework to generate high quality translated images using generative models.

Dipjyoti Paul
Dipjyoti Paul
Ph.D. Student | Research Scientist | Machine Learning | Deep Learning | Audio Signal Processing | Computer Vision | Conversational AI

My research interests include machine learning, deep learning, audio signal processing and image processing.