S&M2174 Research Paper of Special Issue
Published in advance: February 5, 2020
Published: April 10, 2020
Image-similarity-based Convolutional Neural Network for Robot Visual Relocalization [PDF]
Li Wang, Ruifeng Li, Jingwen Sun, Hock Soon Seah, Chee Kwang Quah, Lijun Zhao, and Budianto Tandianus
(Received July 31, 2019; Accepted December 2, 2019)
Keywords: visual relocalization, CNN, image similarity
Convolutional neural network (CNN)-based methods, which train an end-to-end model to regress a six degree of freedom (DoF) pose of a robot from a single red–green–blue (RGB) image, have been developed to overcome the poor robustness of robot visual relocalization recently. However, the pose precision becomes low when the test image is dissimilar to training images. In this paper, we propose a novel method, named image-similarity-based CNN, which considers the image similarity of an input image during the CNN training. The higher the similarity of the input image, the higher precision we can achieve. Therefore, we crop the input image into several small image blocks, and the similarity between each cropped image block and training dataset images is measured by employing a feature vector in a fully connected CNN layer. Finally, the most similar image is selected to regress the pose. A genetic algorithm is utilized to determine the cropped position. Experiments on both open-source dataset 7-Scenes and two actual indoor environments are conducted. The results show that the proposed algorithm leads to better results and reduces large regression errors effectively compared with existing solutions.
Corresponding author: Ruifeng Li, Lijun Zhao
This work is licensed under a Creative Commons Attribution 4.0 International License.
Cite this article
Li Wang, Ruifeng Li, Jingwen Sun, Hock Soon Seah, Chee Kwang Quah, Lijun Zhao, and Budianto Tandianus, Image-similarity-based Convolutional Neural Network for Robot Visual Relocalization, Sens. Mater., Vol. 32, No. 4, 2020, p. 1245-1259.