20240020807.Deep Learning-Based Fusion Techniques for High Resolution, Noise-Reduced, and High Dynamic Range Images with Motion Freezing simplified abstract (apple inc.)

Deep Learning-Based Fusion Techniques for High Resolution, Noise-Reduced, and High Dynamic Range Images with Motion Freezing

Organization Name

apple inc.

Inventor(s)

Marius Tico of Mountain View CA (US)

Tiffany J. Cheng of Cupertino CA (US)

Tanmay Nitin Bichu of Mountain View CA (US)

Deep Learning-Based Fusion Techniques for High Resolution, Noise-Reduced, and High Dynamic Range Images with Motion Freezing - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240020807 titled 'Deep Learning-Based Fusion Techniques for High Resolution, Noise-Reduced, and High Dynamic Range Images with Motion Freezing

Simplified Explanation

The patent application describes electronic devices, methods, and program storage devices for leveraging machine learning to perform high-resolution and low latency image fusion and/or noise reduction.

• An incoming image stream is obtained from an image capture device, consisting of differently-exposed captures received according to a particular pattern.
• When a capture request is received, two or more intermediate assets are generated from images in the incoming image stream and fed into a neural network trained to fuse and/or noise reduce the assets.
• The resultant fused image may have a higher resolution than at least one of the images used to generate the intermediate assets.

Potential Applications

This technology could be applied in:

• Surveillance systems
• Medical imaging
• Satellite imaging

Problems Solved

This technology addresses the following issues:

• Enhancing image resolution
• Reducing noise in images

Benefits

The benefits of this technology include:

• Improved image quality
• Faster image processing
• Enhanced visual clarity

Potential Commercial Applications

Potential commercial applications of this technology include:

• Photography equipment
• Security cameras
• Medical imaging devices

Possible Prior Art

One possible prior art for this technology could be the use of traditional image processing techniques for image fusion and noise reduction.

How does this technology compare to existing image processing methods?

This article does not provide a direct comparison with traditional image processing techniques or other machine learning-based image fusion and noise reduction methods.

What are the limitations of this technology in real-world applications?

The article does not address the potential challenges or limitations that may arise when implementing this technology in practical settings.

Original Abstract Submitted

electronic devices, methods, and program storage devices for leveraging machine learning to perform high-resolution and low latency image fusion and/or noise reduction are disclosed. an incoming image stream may be obtained from an image capture device, wherein the incoming image stream comprises a variety of differently-exposed captures, e.g., ev0 images, ev− images, ev+ images, long exposure images, ev0/ev− image pairs, etc., which are received according to a particular pattern. when a capture request is received, two or more intermediate assets may be generated from images from the incoming image stream and fed into a neural network that has been trained to fuse and/or noise reduce the intermediate assets. in some embodiments, the resultant fused image generated from the two or more intermediate assets may have a higher resolution than at least one of the images that were used to generate at least one of the two or more intermediate assets.