Robust use of semantic segmentation in shallow depth of field rendering

Abstract

This disclosure relates to techniques for the robust usage of semantic segmentation information in image processing techniques, e.g., shallow depth of field (SDOF) renderings. Semantic segmentation may be defined as a process of creating a mask over an image, wherein pixels are segmented into a predefined set of semantic classes. Segmentations may be binary (e.g., a 'person pixel' or a 'non-person pixel') or multi-class (e.g., a pixel may be labelled as: ‘person,’ ‘dog,’ ‘cat,’ etc.). As semantic segmentation techniques grow in accuracy and adoption, it is becoming increasingly important to develop methods of utilizing such segmentations and developing flexible techniques for integrating segmentation information into existing computer vision applications, such as synthetic SDOF renderings, to yield improved results in a wide range of image capture scenarios. In some embodiments, a refinement operation may be employed on a camera device's initial depth, disparity and/or blur estimates that leverages semantic segmentation information.