Similarly, a filtering and subsampling mechanism (104-V) may be configured to filter high spatial frequency content in the vertical direction from the input LE and RE image frames (102-L and 102-R) and vertically subsample the LE and RE image frames (102-L and 102-R) as filtered in the vertical direction into corresponding LE and RE portions. A multiplexer (106-V) may be configured to combine the LE and RE portions in a 3D multiplexed image frame (108-V) in a top-and-bottom format.
The filtering of the LE and RE image frames (102-L and 102-R) may remove all, or a substantial part, of the high spatial frequency content from the input LE and RE image frames (102-L and 102-R) in one of the horizontal and vertical directions. Filtering may be performed with one or more low-pass filters (LPFs) in the filtering and subsampling mechanisms (e.g., 104-H and 104-V). In an example embodiment, filtering as described herein removes or substantially dampens any spatial frequency content in the input images above a threshold frequency that corresponds to a fraction (e.g., one half or another fraction) of a spatial resolution supported by a multi-layer video decoder (e.g., 150) in one of the horizontal and vertical directions.
As used herein, the term “high spatial frequency content” in a spatial direction (horizontal or vertical) may refer to high spatial frequency image details that exist in an input 3D video sequence along the spatial direction. If the removal of the high spatial frequency content in the spatial direction had occurred, downstream devices would not be able to reproduce high resolution image details with filtered image data in the spatial direction.