Real-time automatic conversion of 2-dimensional images or video to 3-dimensional stereo images or video

A computer-implemented method for converting a 2-dimensional (2D) image to 3-dimensional (3D) stereo images comprising:

• determining sharpness values for individual pixels of a 2D input image;
• performing a first region-based sampling of the sharpness values to determine a first plurality of sample sharpness values and a second region-based sampling of the sharpness values to determine a second plurality of sample sharpness values, wherein the first and second region-based samplings are performed at different resolutions;
• scaling the first plurality of sharpness values to generate a first plurality of scaled sharpness values and the second plurality of sharpness values to generate a second plurality of scaled sharpness values;
• smoothing the first plurality of scaled sharpness values to generate a first plurality of smoothed sharpness values and the second plurality of scaled sharpness values to generate a second plurality of smoothed sharpness values;
• determining a first plurality of interpolated sharpness values based on the first plurality of smoothed sharpness values and a second plurality of interpolated sharpness values based on the second plurality of smoothed sharpness values;
• and
• merging the first plurality of interpolated sharpness values and the second plurality of interpolated sharpness values to generate a depth map having a plurality of depth values associated with the 2D input image.

A system for converting a 2-dimensional (2D) image to 3-dimensional (3D) stereo image on a device, comprising:

• a display device configured to present image data;
• one or more processors communicatively coupled to the display device;
• one or more memory stores communicatively coupled to the one or more processors;
• a sharpness measurement module configured to determine sharpness values for individual pixels of a 2D input image;
• at least one sampling module configured to perform a first region-based sampling of the sharpness values to determine a first plurality of sample sharpness values and a second region-based sampling of the sharpness values to determine a second plurality of sample sharpness values, wherein the first and second region-based samplings are performed at different resolutions;
• at least one scaling module configured to scale the first plurality of sharpness values to generate a first plurality of scaled sharpness values and the second plurality of sharpness values to generate a second plurality of scaled sharpness values;
• at least one smoothing module configured to smooth the first plurality of scaled sharpness values to generate a first plurality of smoothed sharpness values and the second plurality of scaled sharpness values to generate a second plurality of smoothed sharpness values;
• at least one interpolation module configured to determine a first plurality of interpolated sharpness values based on the first plurality of smoothed sharpness values and a second plurality of interpolated sharpness values based on the second plurality of smoothed sharpness values;
• and
• a merge module configured to merging the first plurality of interpolated sharpness values and the second plurality of interpolated sharpness values to generate a depth map having a plurality of depth values associated with the 2D input image.

At least one machine readable medium comprising a plurality of instructions that in response to being executed on a computing device, cause the computing device to convert a 2-dimensional (2D) image to 3-dimensional (3D) stereo images by:

• determining sharpness values for individual pixels of a 2D input image;
• performing a first region-based sampling of the sharpness values to determine a first plurality of sample sharpness values and a second region-based sampling of the sharpness values to determine a second plurality of sample sharpness values, wherein the first and second region-based samplings are performed at different resolutions;
• scaling the first plurality of sharpness values to generate a first plurality of scaled sharpness values and the second plurality of sharpness values to generate a second plurality of scaled sharpness values;
• smoothing the first plurality of scaled sharpness values to generate a first plurality of smoothed sharpness values and the second plurality of scaled sharpness values to generate a second plurality of smoothed sharpness values;
• determining a first plurality of interpolated sharpness values based on the first plurality of smoothed sharpness values and a second plurality of interpolated sharpness values based on the second plurality of smoothed sharpness values;
• and
• merging the first plurality of interpolated sharpness values and the second plurality of interpolated sharpness values to generate a depth map having a plurality of depth values associated with the 2D input image.