Display apparatus and image processing method thereof and image processing device thereof with super-resolution for region of interest

An image processing method for display apparatus, comprising:

• receiving a low-resolution image from a host;
• tracking a user's gaze to define a region of interest;
• partitioning the low-resolution image into an original ROI image corresponding to the region of interest and other region image apart from the original ROI image;
• performing a video super-resolution reconstruction on the original ROI image of the low-resolution image to generate a high-definition ROI image;
• and
• stitching the high-definition ROI image and the other region image of the low-resolution image to generate a processed image,
• wherein the region of interest comprises a plurality of sub-regions of interest, the high-definition ROI image comprises a plurality of sub-ROI images, and the image processing method further comprises:
• defining the plurality of sub-regions of interest based on a tracking result of the user's gaze,
• performing the video super-resolution reconstruction on each of the sub-regions of interest in the low-resolution image to generate the plurality of sub-ROI images with different sharpness;
• and
• pasting the plurality of sub-ROI images back to the plurality of sub-regions of interest in the low-resolution image to present a multi-layer progressive sharpness effect between the region of interest and another region in the processed image.

An image processing device, comprising:

• an eye tracking circuit, configured to track a user's gaze;
• a partition circuit, coupled to the eye tracking circuit, wherein the partition circuit receives a low-resolution image from a host, the partition circuit defines a region of interest based on a tracking result of the eye tracking circuit, and the partition circuit partitions the low-resolution image into an original ROI image corresponding to the region of interest and other region image apart from the original ROI image;
• a video super-resolution reconstruction circuit, coupled to the partition circuit to receive the original ROI image, wherein the video super-resolution reconstruction circuit performs a video super-resolution reconstruction on the original ROI image of the low-resolution image to generate a high-definition ROI image;
• and
• a stitching circuit, coupled to the partition circuit to receive the other region image and coupled to the video super-resolution reconstruction circuit to receive the high-definition ROI image, wherein the stitching circuit stitches the high-definition ROI image and the other region image of the low-resolution image to generate a processed image,
• wherein the region of interest comprises a plurality of sub-regions of interest, the high-definition ROI image comprises a plurality of sub-ROI images, and the partition circuit defines the plurality of sub-regions of interest based on the tracking result of the eye tracking circuit,
• the video super-resolution reconstruction circuit performs the video super-resolution reconstruction on each of the sub-regions of interest in the low-resolution image to generate the plurality of sub-ROI images with different sharpness, and
• the stitching circuit pastes the plurality of sub-ROI images back to the sub-regions of interest in the low-resolution image to present a multi-layer progressive sharpness effect between the region of interest and another region in the processed image.