“Smart album: photo filtering by effect detections” by Yu, Liang, Lee, Chen and Ouhyoung

With digital cameras and the storage device technologies are more and more mature, people can easily take large number of pictures in one time. However, some pictures have defect and would be deleted while management, e.g., blur, too dark or too bright images. Moreover, there are also two common problems when we are managing the photos. One is the orientation problem that we always need to rotate the photo if it is not upright, and the other is about duplications that sometimes we will take many shots at the same place with the same viewpoint, but we just keep only one finally. Depends on image analyzing, we make a system “Smart Album” that can automatically filter these photos: blur, abnormal exposure, orientation, and duplication. Thus, users can save lots of time from picking defect pictures up or rotating them.

1. Yongmei Wanga and Hongjiang Zhang 2001. Content-Based Image Orientation Detection with Support Vector Machines. IEEE Workshop on Content-Based Access of Image and Video Libraries (CBAIVL’01), 17–23