Abstract
Image compression is an ever-evolving problem with different approaches coming into prominence at different times. Data analysts are still contemplating about the best approach with compression ratio, visual quality and complexity of the architecture as the main criteria of evaluation. Most recent one is the advancements in machine learning and its applications on image classification. Researchers are currently exploring the possibility of using this advanced computing power to enhance the quality of images with the availability of large datasets. Here, we present a deep learning Convolutional neural network (CNN) implemented and trained for image recognition and image processing tasks, to understand semantics of an image. We achieve higher visual quality in lossy compression by increasing the complexity of an encoder by a slight margin to generate semantic maps of image. The semantic maps are tailored to make the encoder content-aware for a given image. In the proposed work, we present a novel architecture developed specifically for image compression, which generates a semantic map for salient regions so that they can be encoded at higher quality as compared to background regions. Experiments are conducted on the Kodak PhotoCD dataset and the results are compared with other state-of-the-art models. Results of the conducted experiment reveal an increase in compression ratio by more than 17% compared to the referred state-of-the-art model while maintaining the visual quality.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Similar content being viewed by others
References
Agustsson, E., Tschannen, M., Mentzer, F., Timofte, R., Gool, L.V.: Generative adversarial networks for extreme learned image compression. CoRR abs/1804.02958 (2018). http://arxiv.org/abs/1804.02958
Akbari, M., Liang, J., Han, J.: DSSLIC: deep semantic segmentation-based layered image compression. In: ICASSP 2019–2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP), pp. 2042–2046. IEEE (2019)
Cai, C., Chen, L., Zhang, X., Gao, Z.: Efficient variable rate image compression with multi-scale decomposition network. IEEE Trans. Circuits Syst. Video Technol. (2018)
Griffin, G., Holub, A., Perona, P.: Caltech-256 object category dataset. Technical report, 7694, California Institute of Technology (2007). http://authors.library.caltech.edu/7694
Howard, A.G., et al.: MobileNets: efficient convolutional neural networks for mobile vision applications. CoRR abs/1704.04861 (2017). http://arxiv.org/abs/1704.04861
Iandola, F.N., Moskewicz, M.W., Ashraf, K., Han, S., Dally, W.J., Keutzer, K.: SqueezeNet: Alexnet-level accuracy with 50x fewer parameters and \(<\)1mb model size. CoRR abs/1602.07360 (2016). http://arxiv.org/abs/1602.07360
Jiang, F., Tao, W., Liu, S., Ren, J., Guo, X., Zhao, D.: An end-to-end compression framework based on convolutional neural networks. IEEE Trans. Circuits Syst. Video Technol. 28(10), 3007–3018 (2017)
Prakash, A., Moran, N., Garber, S., DiLillo, A., Storer, J.A.: Semantic perceptual image compression using deep convolution networks. CoRR abs/1612.08712 (2016). http://arxiv.org/abs/1612.08712
Santurkar, S., Budden, D., Shavit, N.: Generative compression. In: 2018 Picture Coding Symposium (PCS), pp. 258–262. IEEE (2018)
Selimović, A., Meden, B., Peer, P., Hladnik, A.: Analysis of content-aware image compression with VGG16. In: 2018 IEEE International Work Conference on Bioinspired Intelligence (IWOBI), pp. 1–7. IEEE (2018)
Simonyan, K., Zisserman, A.: Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556 (2014)
Talluri, R., Oehler, K., Barmon, T., Courtney, J.D., Das, A., Liao, J.: A robust, scalable, object-based video compression technique for very low bit-rate coding. IEEE Trans. Circuits Syst. Video Technol. 7(1), 221–233 (1997)
Theis, L., Shi, W., Cunningham, A., Huszár, F.: Lossy image compression with compressive autoencoders. arXiv preprint arXiv:1703.00395 (2017)
Zhao, H., Liao, P.: CAE-ADMM: implicit bitrate optimization via ADMM-based pruning in compressive autoencoders. CoRR abs/1901.07196 (2019). http://arxiv.org/abs/1901.07196
Zhou, B., Khosla, A., Lapedriza, A., Oliva, A., Torralba, A.: Learning deep features for discriminative localization. In: Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, pp. 2921–2929 (2016)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Mahalingaiah, K., Sharma, H., Kaplish, P., Cheng, I. (2020). Semantic Learning for Image Compression (SLIC). In: McDaniel, T., Berretti, S., Curcio, I., Basu, A. (eds) Smart Multimedia. ICSM 2019. Lecture Notes in Computer Science(), vol 12015. Springer, Cham. https://doi.org/10.1007/978-3-030-54407-2_5
Download citation
DOI: https://doi.org/10.1007/978-3-030-54407-2_5
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-54406-5
Online ISBN: 978-3-030-54407-2
eBook Packages: Computer ScienceComputer Science (R0)