Transition between binocular and monocular views

Abstract

The image processing system is designed to generate canvas views with smooth transitions between binocular and monocular views. Initially, the image processing system receives the top / bottom and side images of the scene and calculates offsets to generate a composite side image of the user's left and right views. To achieve a smooth transition between binocular and monocular views, the image processing system first warps the top / bottom images onto the corresponding synthesized side images to generate warped top / bottom images to achieve smooth transitions in shape. The image processing system then warps the distorted top / bottom images onto the corresponding synthesized side images to generate blended images for left and right eye views with blended images. The image processing system creates a canvas view with smooth transitions between binocular and monocular views in image shape and color based on the blended image.

Description

technical field [0001] The present disclosure relates generally to vision systems, and more particularly, to enabling smooth transitions between binocular and monocular views. Background technique [0002] Virtual reality (VR) is becoming more and more popular, and multi-camera systems are often used in VR systems to capture images and / or video to create a virtual scene for a user experiencing the VR system. Images captured by different cameras can be combined into a canvas view for providing a virtual scene, allowing users to experience VR effects. However, problems arise when combining images captured by different cameras to create a composite image. As an example, coarse mixing of images of the same object taken by different cameras can cause double vision in the combined image. Also, when creating a canvas view to simulate a view where the user is looking up and down, the transition between the stereo view and the monocular view may change abruptly or cause a seam betw...

AI Technical Summary

Problems solved by technology

However, problems arise when combining images captured by different cameras to create a composite image

As an example, coarsely blending images of the same object captured by different cameras can cause double vision in the combined image

Also, when creating a canvas view to simulate a view where the user is looking up and down, the transition between the stereo view and the monocular view may change abruptly or cause a seam between the top and side cameras of the binocular view, causing Creates distortions in the view and degrades the user experience of the VR system

Images