ePrints@IIScePrints@IISc Home | About | Browse | Latest Additions | Advanced Search | Contact | Help

CDNet++: Improved Change Detection with Deep Neural Network Feature Correlation

Prabhakar, KR and Ramaswamy, A and Bhambri, S and Gubbi, J and Babu, RV and Purushothaman, B (2020) CDNet++: Improved Change Detection with Deep Neural Network Feature Correlation. In: International Joint Conference on Neural Networks, IJCNN 2020;, 19 July 2020 through 24 July 2020, Virtual, Glasgow; United Kingdom.

[img] PDF
CDNet++.pdf - Published Version
Restricted to Registered users only

Download (5MB) | Request a copy
Official URL: https://doi.org/10.1109/IJCNN48605.2020.9207306


In this paper, we present a deep convolutional neural network (CNN) architecture for segmenting semantic changes between two images. The main objective is to segment changes at the semantic level than detecting background changes, which are irrelevant to the application. The difficulties include seasonal changes, lighting differences, artifacts due to alignment and occlusion. The existing approaches fail to address all the problems together; thus, none of them achieve state-of-the-art performance in three publicly available change detection datasets: VL-CMU-CD 1, TSUNAMI 2 and GSV 2. Our proposed approach is a simple yet effective method that can handle even adverse challenges. In our approach, we leverage the correlation between high-level abstract CNN features to segment the changes. Compared with several traditional and other deep learning-based change detection methods, our proposed method achieves state-of-the-art performance in all three datasets. © 2020 IEEE.

Item Type: Conference Paper
Publication: Proceedings of the International Joint Conference on Neural Networks
Publisher: Institute of Electrical and Electronics Engineers Inc.
Additional Information: Copyright to this article belongs to the IEEE>
Keywords: Convolutional neural networks; Deep learning; Semantics, Change detection; Neural network features; Seasonal changes; Semantic levels; State-of-the-art performance, Deep neural networks
Department/Centre: Division of Interdisciplinary Sciences > Computational and Data Sciences
Date Deposited: 16 Mar 2021 11:01
Last Modified: 16 Mar 2021 11:01
URI: http://eprints.iisc.ac.in/id/eprint/67407

Actions (login required)

View Item View Item