Reading Report: Video Primal Sketch

In this report, we discuss a content-based video coding approach, i.e. video primal sketch. Unlike traditional video coding approaches, video primal sketch seeks to capture different mid-level cues and use them to explain the video data. The algorithm first segmented the video data into two different kinds of regions: explicit region and implicit region, and then model each region by a corresponding mid-level video representations. Specifically, explicit region is represented by sparse coding model, where static or moving primitives, such as moving corners, lines, extinguished feature points, are coded by a generalized dictionary. While implicit region is represented by FRAME model, using spatio-temporal filters to match feature statistics, i.e. histograms.