Abstract With the rapid development of computer vision technology, increasingly more focus has been put on image recognition. More specifically, a sketch is an important hand-drawn image that is garnering increased attention. Moreover, as handheld devices such as tablets, smartphones, etc. have become more popular, it has become increasingly more convenient for people to hand-draw sketches using this equipment. Hence, sketch recognition is a necessary task to improve the performance of intelligent equipment. In this paper, a sketch recognition learning approach is proposed that is based on the Visual Geometry Group16 Convolutional Neural Network (VGG16 CNN). In particular, in order to diminish the effect of the number of sketches on the learning method, we adopt a strategy of increasing the quantity to improve the diversity and scale of sketches. Initially, sketch features are extracted via the pretrained VGG16 CNN. Additionally, we obtain contextual features based on the traverse stroke scheme. Then, the VGG16 CNN is trained using a joint Bayesian method to update the related network parameters. Moreover, this network has been applied to predict the labels of input sketches in order to automatically recognize the label of a sketch. Last but not least, related experiments are conducted, and the comparison of our method with the state-of-the-art methods is performed, which shows that our approach is superior and feasible.

References

[1] Bay H., Tuytelaars T., and Gool L., “SURF: Speeded Up Robust Features,” Computer Vision and Image Understanding, vol. 110, no. 3, pp. 346-359, 2006.

[2] Chen D., Cao X., Wang L., Wen F., and Sun J., “Bayesian Face Revisited: A Joint Formulation,” in Proceedings of European Conference on Computer Vision, Florence, pp. 566-579, 2012.

[3] Donahue J., Jia Y., Vinyals O., Hoffman J., Zhang N., Tzeng E., and Darrell T., “DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition,” in Proceedings of International Conference on Machine Learning, pp. 647-655, 2014.

[4] Eitz M., Hays J., and Alexa M., “How Do Humans Sketch Objects?,” ACM Transactions on Graphics, vol. 31, no. 4, pp. 1-10, 2012.

[5] Girshick R., Donahue J., Darrell T., Malik J., and Berkeley U., “Rich Feature Hierarchies for Accurate Object Detection and Semantic Segmentation,” in Proceedings of Computer Vision and Pattern Recognition, Columbus, pp. 580-587, 2014.

[6] Jabal M., Rahim M., Othman N., and Jupri Z., “A Comparative Study on Extraction and Recognition Method of CAD Data from CAD Drawings,” in Proceedings of International Conference on Information Management and Engineering. IEEE Computer Society, Kuala Lumpur, pp. 709-713, 2009.

[7] Li B., Lu Y., Godil A., and Schreck T., “SHREC'13 Track: Large Scale Sketch-Based 3D Shape Retrieval,” in Proceedings of Eurographics Workshop on 3D Object Retrieval. Eurographics Association, Girona, pp. 89-96, 2013.

[8] Liang S., Zhao L., Wei Y., and Jia J., “Sketch- Based Retrieval Using Content-Aware Hashing,” in Proceedings of Pacific Rim Conference on Multimedia, Kuching, pp. 133-142, 2014.

[9] Lowe D., “Distinctive Image Features from Scale-Invariant Keypoints,” International Journal of Computer Vision, vol. 60, no. 2, pp. 91-110, 2004.

[10] Lu T., Tai C., Su F., and Cai S., “A New Recognition Model for Electronic Architectural Drawings,” Computer-Aided Design, vol. 37, no. 10, pp. 1053-1069, 2005.

[11] Ma C., Yang X., Zhang C., Ruan X., and Yang M., “Sketch Retrieval Via Dense Stroke Features,” Image and Vision Computing,vol. 46, no. 2, pp. 64-73, 2016.

[12] Mikolajczyk K. and Schmid C., “Scale and Affine Invariant Interest Point Detectors,” International Journal of Computer Vision, vol. 60, no. 1, pp. 63-86, 2004.

[13] Razavian A., Azizpour H., Sullivan J., and Carlsson S., “CNN Features Off-the-Shelf: An Astounding Baseline for Recognition,” in Proceedings of Computer Vision and Pattern Recognition, Columbus, pp. 806-813, 2014.

[14] Reddy N., Rao M., and Satyanarayana C., “Novel Face Recognition System by the Combination of Multiple Feature Descriptors” The International Arab Journal of Information Technology, vol. 16, no. 4, pp. 669-676, 2019.

[15] Schneider R. and Tuytelaars T., “Sketch Classification and Classification-Driven Analysis Using Fisher Vectors,” ACM Transactions on Graphics, vol. 33, no. 6, pp. 174-184, 2014.

[16] Simonyan K. and Zisserman A., “Very Deep Convolutional Networks for Large-Scale Image Recognition,” in Proceedings of International Conference on Learning Representations, San Diego, pp. 1-14, 2015. Training Convolutional Neural Network for Sketch Recognition on Large-Scale Dataset 89

[17] Wang F., Kang L., and Li Y., “Sketch-based 3D shape Retrieval Using Convolutional Neural Networks,” in Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition, Boston, pp. 1875-1883, 2015.

[18] Yi L., Timothy M., Song Y., and Gong S., “Free- Hand Sketch Recognition By Multi-Kernel Feature Learning,” Computer Vision and Image Understanding, vol. 137, pp. 1-11, 2015.

[19] Yu Q., Yang Y., Liu F., and Song Y., “Sketch-a- Net: A Deep Neural Network that Beats Humans,” International Journal of Computer Vision, vol. 122, no. 3, pp. 411-425, 2016.

[20] Yu Q., Yang Y., Song Y., and Xiang T.,“Sketch- a-Net that Beats Humans,” in Proceedings of British Machine Vision Conference, Swansea, pp. 101-112, 2015.

[21] Zitnick C. and Parikh D., “Bringing Semantics into Focus Using Visual Abstraction,” IEEE Conference on Computer Vision and Pattern Recognition, Portland, pp. 3009-3016, 2013. Wen Zhou received the Ph.D. degree from School of Software Engineering, Tongji University in 2018. Since2018, he has been in the School of Computer and Information, Anhui Normal University, China, where he is currently a lecturer, IEEE Member, Member of Chinese Computer Federation (CCF). His research interests include Virtual Reality, Sketch-based Retrieval and Machine Learning etc. Jinyuan Jia received the Ph.D. degree from TheHong Kong University of Science and Technologyin 2004. Since 2007, he has been with Tongji University, Shanghai, China, where he is currently a Professor.His research interests include computer graphics, Web3D, mobile VR,etc. He is an ACM Member, and a Senior Member of the Chinese Computer Federation.