18558611. SUPER RESOLUTION BASED ON SALIENCY simplified abstract (QUALCOMM Incorporated)
Contents
SUPER RESOLUTION BASED ON SALIENCY
Organization Name
Inventor(s)
Weixing Wan of Baoshan District, Shanghai (CN)
SUPER RESOLUTION BASED ON SALIENCY - A simplified explanation of the abstract
This abstract first appeared for US patent application 18558611 titled 'SUPER RESOLUTION BASED ON SALIENCY
Simplified Explanation: The patent application describes systems and techniques for image processing, where an imaging system can enhance the resolution of specific regions in an input image based on their saliency.
Key Features and Innovation:
- Imaging system can identify salient regions in an input image.
- Utilizes different processes to increase the resolution of these regions.
- Outputs an image with modified high-resolution regions.
Potential Applications: This technology can be used in various fields such as medical imaging, satellite imaging, surveillance, and photography.
Problems Solved: This technology addresses the issue of enhancing image resolution selectively based on the importance of different regions.
Benefits:
- Improved image quality in specific regions.
- Enhanced visual clarity and detail.
- Efficient use of computational resources.
Commercial Applications: The technology can be applied in industries such as healthcare, security, entertainment, and remote sensing for commercial purposes.
Prior Art: Researchers can explore prior art related to machine learning-based image processing systems and super resolution techniques.
Frequently Updated Research: Stay updated on advancements in machine learning algorithms for image processing and super resolution techniques.
Questions about Image Processing: 1. How does the imaging system determine the saliency of different regions in an input image? 2. What are the potential limitations of using machine learning for increasing image resolution?
Original Abstract Submitted
Systems and techniques are described for image processing. For instance, an imaging system can obtain an input image with a first region and a second region, both at a first resolution. The imaging system can determine that the first region is more salient than the second region (e.g., based on a saliency map mapping saliency values to pixels of the input image). The imaging system can use a first process (e.g., using a trained network, such as of a machine learning super resolution system) to modify the first region to increase the first resolution to a second resolution. The imaging system can use a second process (e.g., based on an interpolation process) to modify the second region to increase the first resolution of the second region to the second resolution. The imaging system can generate and/or output an output image including the modified first region and the modified second region.
