18451648. COMPOSITE IMAGE GENERATION BASED ON IMAGE SEGMENTATION (QUALCOMM Incorporated)

COMPOSITE IMAGE GENERATION BASED ON IMAGE SEGMENTATION

Organization Name

QUALCOMM Incorporated

Inventor(s)

Dharanya Vanchinathan of San Diego CA (US)

Shizhong Liu of San Diego CA (US)

Narayana Karthik Ravirala of San Diego CA (US)

Kevin Yao Han of San Diego CA (US)

Deepak Subramani Velumani of San Diego CA (US)

Weiliang Liu of San Diego CA (US)

COMPOSITE IMAGE GENERATION BASED ON IMAGE SEGMENTATION

This abstract first appeared for US patent application 18451648 titled 'COMPOSITE IMAGE GENERATION BASED ON IMAGE SEGMENTATION

Original Abstract Submitted

Systems and techniques are provided for processing image data. First image data of a scene can be obtained with a first zoom level and including at least a foreground portion and a background portion. A user input can be received indicative of an adjustment to increase or decrease a zoom level of the foreground relative to the background, corresponding to a second zoom level greater or less than the first zoom level. A second image data of the scene can be obtained, based on the adjustment and using the second zoom level, to include an adjusted foreground portion associated with the second zoom level. The adjusted foreground portion can be segmented from the second image data of the scene. A composite image can be generated based on combining the segmented foreground portion from the second image data with at least a portion of the first image data.