Abstract
In visual recognition, the key to the performance improvement of ResNet is the success in establishing the stack of deep sequential convolutional layers using identical mapping by a shortcut connection. It results in multiple paths of data flow under a network and the paths are merged with the equal weights. However, it is questionable whether it is correct to use the fixed and predefined weights at the mapping units of all paths. In this paper, we introduce the active weighted mapping method which infers proper weight values based on the characteristic of input data on the fly. The weight values of each mapping unit are not fixed but changed as the input image is changed, and the most proper weight values for each mapping unit are derived according to the input image. For this purpose, channel-wise information is embedded from both the shortcut connection and convolutional block, and then the fully connected layers are used to estimate the weight values for the mapping units. We train the backbone network and the proposed module alternately for a more stable learning of the proposed method. Results of the extensive experiments show that the proposed method works successfully on the various backbone architectures from ResNet to DenseNet. We also verify the superiority and generality of the proposed method on various datasets in comparison with the baseline.
Similar content being viewed by others
Change history
07 October 2021
A Correction to this paper has been published: https://doi.org/10.1007/s11042-021-11538-z
References
Duda RO, Hart PE, Stork DG (2000) Pattern classification, 2nd edn. Wiley-Interscience, New York
Han D, Kim J, Kim J (2017) Deep pyramidal residual networks. IEEE Conf Comput Vision Pattern Recogn, pp 6307–6315
Hassaballah M, Awad AI (2020) Deep learning in computer vision: principles and applications. CRC Press
He K, Zhang X, Ren S, Sun J (2016) Deep residual learning for image recognition. IEEE Conf Comput Vision Pattern Recogn, pp 770–778
He K, Zhang X, Ren S, Sun J (2016) Identity mappings in deep residual networks. European Conf Comput Vision, pp 630–645
Hu J, Shen L, Sun G (2018) Squeeze-and-excitation networks. IEEE Conf Comput Vision Pattern Recogn
Huang G, Liu Z, van der Maaten L, Weinberger K (2017) Densely connected convolutional networks. IEEE Conf Comput Vision Pattern Recogn, pp 2261–2269
Huang G, Sun Y, Liu Z, Sedra D, Weinberger KQ (2016) Deep networks with stochastic depth. European Conf Comput Vision 4:646–661
Hwang W (2018) Sequential convolutional residual network for image recognition. IEICE Trans Inform Sys E101-D(4):1213–1216
Krizhevsk A, Sulskever I, Hinton GE (2012) Imagenet classification with deep convolutional neural networks. Neural Inform Process Sys 25:1097–1105
Krizhevsky A (2009) Learning multiple layers of feature from tiny images. Tech. report
Larsson G, Maire M, Shakhnarovich G (2017) Fractalnet: ultra-deep neural networks without residuals. International Conference on Learning Representation
Liu L, Wang L, Liu X (2011) In defense of soft-assignment coding. IEEE Conf Comput Vision, pp 2486–2493
Martinez A, Kak A (2001) Pca versus lda. IEEE Tran Pattern Recogn Machine Intell 23(2):228–233
Pytorch. http://pytorch.org
Russakovsky O, Deng J, Su H, Krause J, Satheesh S, Ma S, Huang Z, Karpathy A, Khosla A, Bernstein M, Berg AC, Fei-Fei L (2015) Imagenet large scale visual recognition challenge. Int J Comput Vis 115(3):211–252
Simonyan K, Zisserman A (2015) Very deep convolutional networks for large-scale image recognition. International Conference on Learning Representations
Singh RD, Mittal A, Bhatia RK (2019) 3d convolutional neural network for object recognition: a review. Multimedia Tools and Applications 78(12)
Srivastava RK, Greff K, Schmidhuber J (2015) Highway networks. arXiv:1505.00387
Sun X, Xv H, Dong J, Zhou H, Chen C, Li Q (2020) Few-shot learning for domainspecific fine-grained image classification. IEEE Transactions on Industrial Electronics
Sun Y, Xue B, Zhang M, Yen GG, Lv J (2020) Automatically designing cnn architectures using the genetic algorithm for image classification. IEEE Transactions on Cybernetics
Szegedy C, Liua W, Jia Y, Sermanet P, Reed S, Anguelov D, Erhan D, Vanhoucke V, Rabinovich A (2015) Going deeper with convolutions. IEEE Conf Comput Vision Pattern Recogn, pp 1–9
Tan M, Le Q (2019) Efficientnet: rethinking model scaling for convolutional neural networks. International Conference on Machine Learning
Veit A, Belongie S (2018) Convolutional networks with adaptive inference graphs. European Conf Comput Vision
Veit A, Wilber M, Belongie S (2016) Residual networks behave like ensembles of relatively shallow networks. Neural Information Processing Systems 29:550–558
Wang F, Jiang M, Qian C, Yang S, Li C, Zhang H, Wang X, Tang X (2017) Residual attention network for image classification. arXiv:1704.06904
Wang X, Yu F, Dou ZY, Darrell T, Gonzalez JE (2018) Skipnet: learning dynamic routing in convolutional networks. European Conf Comput Vision
Woo S, Park J, Lee J, Kweon I (2018) CBAM: convolutional block attention module. European Conf Comput Vision
Xie S, Girshick R, Dollar P, Tu Z, He K (2017) Aggregated residual transformations for deep neural networks,. IEEE Conf Comput Vision Pattern Recogn
Zagoruyko S, Komodakis N (2016) Wide residual network. British Machine Vision Conference, pp 87.1–87.12
Zoph B, Le Q (2017) Neural architecture search with reinforcement learning. International Conference on Learning Representations
Acknowledgements
This work was supported by a Research and Development project, Enabling a System for Sharing and Disseminating Research Data of Korea Institute of Science and Technology (KISTI), South Korea, under Grant K-20-L01-C04-S01. The corresponding author is Dr. Wonjun Hwang.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
The original online version of this article was revised: The given name of the first author was misspelled as “Hyungho” and the images of Figs. 7, 8, 9and 10 were interchanged.
Rights and permissions
About this article
Cite this article
Jung, H., Lee, R., Lee, SH. et al. Active weighted mapping-based residual convolutional neural network for image classification. Multimed Tools Appl 80, 33139–33153 (2021). https://doi.org/10.1007/s11042-020-09808-3
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11042-020-09808-3