This paper shows that using synthetic images for violence detection is a feasible option for training deep models, they aid in quick convergence and perform better.

This paper solves the textual deblurring problem, In this paper we propose a new loss function, we provide empirical evaluation of the design choices based on which a memory friendly CNN model is proposed, that performs better then the state of the art CNN method.

A document by Muhammad Shahroz Nadeem . Click on the document to view its contents.