Automatic Colorization of Black and White Images using Deep Learning

Abstract
Authors
Keywords
Conclusion
References

The main aim of our paper is to give an overall idea about how a grayscale image can be converted into a colorful image with the colorization problem and how it can further be used to color a video. To achieve artifact-free quality this generally requires manual reconciliation and therefore considered as a very strenuous problem . A cautious selection of colorful allusion images are generally required for the process. Far from the preceding methods, this paper aims at a high grade fully unmanned colorization method and also attempt to apply this concept to images obtained from video sequences. The recent achievements in deep learning approaches is the inspiration behind this paper, that focuses on reformulation of the problem of colorization so as we can employ the deep learning approaches promptly and that this technique can be applied on to the videos. Our proposed method is a fully automated process. To our best apprehension, no prevailing papers or research studies label this issue of using deep learning techniques to colorize videos.

Published In : IJCSN Journal Volume 8, Issue 2

Date of Publication : April 2019

Pages : 125-131

Figures :08

Tables : --

Sindhuja Kotala : is currently pursuing B.E III-Year, CSE at Stanley college of Engineering and Technologies affiliated to Osmania University. Her area of interests are Artificial Intelligence, Machine Learning , Deep Learning, Predictive Analysis.

Srividya Tirumalasetti : is currently pursuing B.E IIIYear, CSE at Stanley college of Engineering and Technologies affiliated to Osmania University. Her area of interests are Artificial Intelligence, Machine Learning.

Vudaru Nemitha : is currently pursuing B.E III-Year, CSE at Stanley college of Engineering and Technologies affiliated to Osmania University.

Swapna Munigala : CSE Department, Osmania University- Stanley college of Engineering and Technology for Women, Asst Prof CSE Dept, Hyderabad, Telangana 500036, India

Deep learning, Convoluted neural networks, Machine learning

We have presented a method of fully automatic colorization of unique greyscale images combining stateof- the-art CNN techniques[5]. Using color representation and the right loss function, we have represented that the method is capable of producing a plausible and vibrant colorization of certain parts of images that has properties which may be applied to video sequences also. Our model does very well with the animals like cats and dogs because the dataset we chose i.e., ImageNet consists large amount of pictures of these animals[12]. Even the outdoor scenes turnout very good with our model. The model also captures notion of sunset and paints it orange.

[1] Iizuka, Satoshi, Edgar Simo-Serra, and Hiroshi Ishikawa. "Let there be color!: joint end-to-end learning of global and local image priors for automatic image colorization with simultaneous classification." ACM Transactions on Graphics (TOG) 35.4 (2016): 110. [2] Cheng, Zezhou, Qingxiong Yang, and Bin Sheng. "Deep colorization." Proceedings of the IEEE International Conference on Computer Vision. 2015. [3] Larsson, Gustav, Michael Maire, and Gregory Shakhnarovich. "Learning representations for automatic colorization." European Conference on Computer Vision. Springer, Cham, 2016. [4] Li, Bo, et al. "Depth and surface normal estimation from monocular images using regression on deep features and hierarchical CRFs." Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition. 2015. [5] Zhang, Richard, et al. "Real-time user-guided image colorization with learned deep priors." arXiv preprint arXiv:1705.02999 (2017). [6] Deshpande, Aditya, Jason Rock, and David Forsyth. "Learning large-scale automatic image colorization." Proceedings of the IEEE International Conference on Computer Vision. 2015. [7] Limmer, Matthias, and Hendrik PA Lensch. "Infrared colorization using deep convolutional neural networks." Machine Learning and Applications (ICMLA), 2016 15th IEEE International Conference on. IEEE, 2016. [8] Deshpande, Aditya, Jason Rock, and David Forsyth. "Learning large-scale automatic image colorization." Proceedings of the IEEE International Conference on Computer Vision. 2015. [9] C. Dong, C. C. Loy, K. He,and X. Tang. Learning a deep convolutional network for image super resolution. In ECCV, pages 184-199. Springer, 2014. [10] Z. Yan, H. Zhang, B. Wang, S. Paris, and Y. Yu. Automatic photo adjustment using deep learning. arXiv preprint arXiv:1412.7725, 2014. [11] Charpiat, Guillaume, Matthias Hofmann, and Bernhard Schölkopf. "Automatic image colorization via multimodal predictions." European conference on computer vision. Springer, Berlin, Heidelberg, 2008. [12] Zhang, Richard, Phillip Isola, and Alexei A. Efros. "Colorful image colorization." European Conference on Computer Vision. Springer, Cham, 2016. [13] LeCun, Yann, Yoshua Bengio, and Geoffrey Hinton. "Deep learning." nature 521.7553 (2015): 436. [14] Krizhevsky, Alex, Ilya Sutskever, and Geoffrey E. Hinton. "Imagenet classification with deep convolutional neural networks." Advances in neural information processing systems. 2012. [15] Marr, David. "Artificial intelligence-a personal view." Artificial Intelligence 9.1 (1977): 37-48. [16] Nasrabadi, Nasser M. "Pattern recognition and machine learning." Journal of electronic imaging 16.4 (2007): 049901.