Abstract

Online personalized advertisements on a smartphone have shown a great impact on both user experience and advertiser income. This type of personalized advertising is possible due to existence of easily traceable features when user is online. Online advertising agencies such as Google AdSense can determine appropriate ads for a particular user based on their behavior on the internet. Therefore, these advertising agencies inherently depend long interactions between user and the device to get a decent ad recommendation. This paper focuses on interactions of users with electronic devices which are very short and need-based. Examples of these devices would be gaming arenas, selfie stations in malls or self check-in booths at the airport. The paper throughout considers these types of interactions as ”offline” since there is no way to track user’s behavior here like it is possible in ”online” scenario. Main objective of the paper is to discuss challenges in recommending ads in offline interaction scenario and develop methods to overcome these challenges. Finally, the paper presents a method to recommend ads using fashion based features with the help of computer vision and demonstrates its working.

References

• ARM NEON instruction set. http:// infocenter:arm:com/help/index:jsp?topic= /com:arm:doc:dht0002a/ch01s03s03:html. Accessed: 2020-06-13.
• BeagleBone AI. https://beagleboard:org/ai. Accessed: 2020-06-13.
• Odroid C2. https://wiki:odroid:com/odroid-c2/ odroid-c2. Accessed: 2020-06-13. Pine A64-LTS. https://www:pine64:org/devices/ single-board-computers/pine-a64-lts/. Accessed: 2020-06-13.
• Raspberry Pi 4. https://www:raspberrypi:org/ products/raspberry-pi-4-model-b/ specifications/. Accessed: 2020-06-13.
• Rock64. https://www:pine64:org/devices/single- board-computers/rock64/. Accessed: 2020-06-13.
• Zhe Cao, Tomas Simon, Shih-En Wei, and Yaser Sheikh. Realtime multi-person 2d pose estimation using part affinity fields. In The IEEE Conference on Computer Vision and Pattern Recognition (CVPR), July 2017.
• K. He, X. Zhang, S. Ren, and J. Sun. Deep residual learning for image recognition. In 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR), pages 770– 778, June 2016.
• Khalid Saleh. Effectiveness of online advertising statistics and trends. [Online; Accessed 14-11-2019].
• Alex Krizhevsky, Ilya Sutskever, and Geoffrey E Hinton. Imagenet classification with deep convolutional neural networks. In F. Pereira, C. J. C. Burges, L. Bottou, and K. Q. Weinberger, editors, Advances in Neural Information Processing Systems 25, pages 1097–1105. Curran Associates, Inc., 2012.
• Yann LeCun, Yoshua Bengio, and Geoffrey Hinton. Deep learning. Nature, 521:436 EP –, May 2015.
• Daniil Osokin. Real-time 2d multi-person pose estimation on CPU: lightweight openpose. CoRR, abs/1811.12004, 2018.
• Mark Sandler, Andrew Howard, Menglong Zhu, Andrey Zhmoginov, and Liang-Chieh Chen. MobileNetV2: Inverted Residuals and Linear Bottlenecks. arXiv e-prints, page arXiv:1801.04381, January 2018.
• Rui Yang Bin Luo Jin Tang Xiao Wang, Shaofei Zheng. Pedestrian attribute recognition: A survey. arXiv preprint arXiv:1901.07474, 2019.