Mobile phone panoramic shooting and synthesizing method based on end-side deep learning

Abstract

The invention discloses a mobile phone panoramic shooting and synthesizing method based on end-side deep learning. The method mainly utilizes an end-side reasoning framework and a deep neural network model to carry out image splicing, image enhancement deblurring and image content complementation on front and rear shooting contents of a mobile phone, and the most real shooting and synthesizing scene is restored, so that the shooting visual field range can be expanded, the observation visual angle can be changed, a sense of immediacy is brought to a user, no professional panoramic shooting equipment is needed, and shooting can be carried out by turning on the mobile phone anytime and anywhere. The threshold of VR shooting and synthesis is reduced, and each user can carry out VR panoramic shooting by using the mobile phone.

Description

technical field [0001] The present invention relates to the field of virtual reality technology, and more specifically, relates to a mobile phone panoramic shooting and synthesis method based on end-side deep learning. Background technique [0002] With the development of virtual reality (VR), it mainly refers to 360° video, also known as panoramic video. The combination of VR is the development trend of VR in the future. It has a wide range of applications, such as sports events, variety shows, news scenes, education and medical care, games and e-sports, etc. At the same time, the emergence of the end-test reasoning engine enables the deep learning model to be directly deployed on the mobile phone, making it possible to perform real-time panoramic shooting and synthesis through the mobile phone. Then, combining the immersive experience brought by VR with the low cost and The combination of low threshold has become the focus of current research. [0003] It is mentioned i...

AI Technical Summary

Problems solved by technology

However, the parameters of the front and rear cameras of the mobile phone are different, so the collected image data will have different brightness, different resolution, and different color saturation, which will cause the two captured images to have certain differences, which will cause artifacts during the splicing process. , and the existence of artifacts will greatly destroy the user's visual experience

And although the mobile phone lens has an ultra-wide-angle mode, the shooting range of the mobile phone’s front and rear cameras is still not enough to cover the 360° panorama after stitching, and there are missing content in the stitching screen

Images