Face portrait synthesis method based on greedy search

Abstract

The invention discloses a face portrait synthesis method based on greedy search. The face portrait synthesis method comprises the following steps: dividing a dictionary training picture sample set A, synthesizing a training picture sample set Bp, synthesizing a training portrait sample set Bs and testing a picture sample set; blocking so as to obtain a dictionary training picture block set SA, synthesizing a training picture block set Sp, synthesizing a training portrait block set Ss and testing a picture block set; learning a training picture block characteristic dictionary Dp from the dictionary training picture block set SA; solving a sparse representation set Cp corresponding to Sp and testing sparse representation corresponding to the picture block S by using Dp; performing greedy search in Cp by using sparse representation so as to find neighbors; establishing a Markov random field model by using the searched neighbor portrait-picture blocks so as to synthesize portraits. Compared with a conventional method, the face portrait synthesis method does not need to limit the background of a tested picture, is capable of combining non-face components in the tested picture, and can be applied to criminal investigation and digital entertainment.

Description

technical field [0001] The invention belongs to the technical field of image processing, and further relates to a method for synthesizing human face portraits in the technical fields of pattern recognition and computer vision, which can be used for criminal investigation and digital entertainment. Background technique [0002] Face portrait synthesis has important practical applications, such as supporting animation production and assisting the police in solving crimes. When a case occurs, the police will generally draw a portrait of the suspect based on the description of the witnesses at the scene, and then search and identify it in the police's face photo database. Since photos and portraits belong to two different modalities, there are large differences. Therefore, before the recognition, the photos in the police face photo database are converted into portraits, and then the suspect's portrait is identified in the synthetic portrait database. At present, social media is...

AI Technical Summary

Problems solved by technology

However, this method has the following disadvantages: the fixed number of neighbors makes the synthetic portrait unclear

The disadvantage of this method is: because each photo block to be synthesized finally selects only one training image block for image synthesis, resulting in block effects and missing details in the synthesized image

The disadvantages of this method are: the linear superposition of detail information makes the composite image unnatural, and the quality of the composite image mostly depends on the initial image synthesis method

[0007] The above three types of portrait synthesis methods are only applicable to the situation where the background of the test photo is the same as that of the training set photo. When the background of the test photo is different from the background of the training set photo, the background of the image synthesized by the above three types of methods will have noise.

In addition, when there are non-face parts such as glasses and hairpins in the test photos, but these non-face parts do not exist in the photos of the training set, the corresponding portraits synthesized by the above three methods will lose the non-face parts in the test photos. face parts

Images