Continuous digital zooming using local self-similarity-based super-resolution for an asymmetric dual camera system

Area of Science:

• Computer Vision
• Image Processing
• Digital Photography

Background:

• Asymmetric dual camera systems capture images with varying resolutions.
• Existing single-image super-resolution (SR) methods have limitations.

Purpose of the Study:

• To develop an SR algorithm for digital zooming using local self-similarity.
• To enhance low-resolution wide-view images by leveraging information from tele-view images.

Main Methods:

• Registration of optically zoomed images to wide-view images.
• Restoration of central and boundary regions of the zoomed wide-view image.
• Fusion of restored image regions.

Main Results:

• The proposed method effectively restores low-resolution wide-view images.
• Significantly improved high-resolution images are achieved compared to existing SR methods.
• Leverages high-frequency components from optically zoomed images.

Conclusions:

• The developed SR algorithm offers superior performance for digital zooming with asymmetric dual cameras.
• This method enhances image quality by utilizing the complementary information from different camera views.