Simple Domain Adaptation for CAD based Object Recognition

Kripasindhu Sarkar, Didier Stricker

Abstract

We present a simple method of domain adaptation between synthetic images and real images - by high quality rendering of the 3D models and correlation alignment. Using this method, we solve the problem of 3D object recognition in 2D images by fine-tuning existing pretrained CNN models for the object categories using the rendered images. Experimentally, we show that our rendering pipeline along with the correlation alignment improve the recognition accuracy of existing CNN based recognition trained on rendered images - by a canonical renderer - by a large margin. Using the same idea we present a general image classifier of common objects which is trained only on the 3D models from the publicly available databases, and show that a small number of training models are sufficient to capture different variations within and across the classes.

Download


Paper Citation


in Harvard Style

Sarkar K. and Stricker D. (2019). Simple Domain Adaptation for CAD based Object Recognition.In Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM, ISBN 978-989-758-351-3, pages 429-437. DOI: 10.5220/0007346504290437


in Bibtex Style

@conference{icpram19,
author={Kripasindhu Sarkar and Didier Stricker},
title={Simple Domain Adaptation for CAD based Object Recognition},
booktitle={Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,},
year={2019},
pages={429-437},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0007346504290437},
isbn={978-989-758-351-3},
}


in EndNote Style

TY - CONF

JO - Proceedings of the 8th International Conference on Pattern Recognition Applications and Methods - Volume 1: ICPRAM,
TI - Simple Domain Adaptation for CAD based Object Recognition
SN - 978-989-758-351-3
AU - Sarkar K.
AU - Stricker D.
PY - 2019
SP - 429
EP - 437
DO - 10.5220/0007346504290437