The second part of the practice review is about photo recommendation.
Photo Recommendation for SmartPhone
Regarding the storage of electronic photos, I investigated the photo arrangement of some mobile phone albums and the way of recommending photos. Most mobile phone albums are arranged in chronological order. Take the Apple mobile phone as an example, mobile phone albums include ordinary photos and favourite photos which are manually added by users. With tens of thousands of photos, the manual organization is a difficult task (Kucer & Messinger, 2018).
Therefore, systems of photo sorting and recommendation were born. Early photo aesthetic algorithms mainly used to judge the quality of the photo to predict whether the photo is aesthetic (Kucer & Messinger, 2018). In this project, the image quality and aesthetic level were judged. The project first selects multiple factors that determine the quality and beauty of the photo. After that, a large number of photos were imported into MobileNet CNN architecture to train the accuracy of the model. Finally, a model that can better judge the beauty of the photo was completed (Howard et al., 2017).
These algorithm models are applied to mobile phones to select beautiful photos in albums and recommend them to users. Some photos will be made into videos and displayed to users together to increase the browsing rate of beautiful photos in the album. Most shelf life reminding projects allow users to manually enter the shelf life by scanning codes. Regarding mobile phone photo album recommendation, it is the needs of many users to recommend beautiful photos in mobile phone photo albums, and there are many ways to recommend them currently. Therefore, combined with the above research, the project can deeply explore how to use the expiry date to help users look back at the wonderful moments in the mobile photo album.
Reference
Kucer, M., & Messinger, D. W. (2018). Aesthetic Inference for Smart Mobile Devices. 2018 IEEE Winter Conference on Applications of Computer Vision (WACV).doi:10.1109/wacv.2018.00196
Howard, A. G., Zhu, M., Chen, B., Kalenichenko, D., Wang, W., Weyand, T., Andreetto, M., & Adam, H. (2017). Mobilenets: Efficient convolutional neural networks for mobile vision applications. CoRR, abs/1704.04861, 2017.
Comments