Current state of the art object recognition architectures achieve impressive performance but are typically specialized for a single depictive style (e.g. photos only, sketches only). In this paper, we present SwiDeN : our Convolutional Neural Network (CNN) architecture which recognizes objects regardless of how they are visually depicted (line drawing, realistic shaded drawing, photograph etc.). In SwiDeN, we utilize a novel `deep' depictive style-based switching mechanism which appropriately addresses the depiction-specific and depiction-invariant aspects of the problem. We compare SwiDeN with alternative architectures and prior work on a 50-category Photo-Art dataset containing objects depicted in multiple styles. Experimental results show that SwiDeN outperforms other approaches for the depiction-invariant object recognition problem.
@article{arxiv.1607.08764,
title = {SwiDeN : Convolutional Neural Networks For Depiction Invariant Object Recognition},
author = {Ravi Kiran Sarvadevabhatla and Shiv Surya and Srinivas S S Kruthiventi and Venkatesh Babu R},
journal= {arXiv preprint arXiv:1607.08764},
year = {2016}
}
Comments
Accepted at ACMMM 2016. The first two authors contributed equally. Code and models at https://github.com/val-iisc/swiden