Detail preserving denoiser for short exposure images

Abstract

A method includes obtaining a noisy training image and a ground truth image and converting the noisy training image into a first plurality of color channels. The method also includes generating, using an artificial intelligence (AI)-based denoiser, denoised images from the first plurality of color channels. Each denoised image corresponds to a respective color channel of the first plurality of color channels. The method further includes generating a noisy ground truth image from the ground truth image and converting the noisy ground truth image into a second plurality of color channels. In addition, the method includes determining a loss based on a comparison between the denoised images and the second plurality of color channels and adapting weights of the AI -based denoiser based on the loss determined based on the comparison between the denoised images and the second plurality of color channels.

Description

DETAIL PRESERVING DENOISER FOR SHORT EXPOSURE IMAGES

[0001] This disclosure relates generally to image processing. More specifically, this disclosure relates to a detail preserving denoiser for short exposure images.

[0002] Denoising is a useful or important operation in an image signal processing (ISP) pipeline or other image processing pipeline. Image noise is a phenomenon that is particularly noticeable in low light or short exposures, often manifesting as speckles or blur in an image.　Image sensors introduce various types of noise. For example, shot noise refers to noise caused by the random nature of quantum interactions of photons with atoms of a sensor. Dark current noise refers to noise caused by thermally-generated electronics. Quantization noise refers to noise caused by inaccuracies in the process of "counting" electrical charges generated by incoming photons.　Noise reduction algorithms can be employed to suppress these imperfections while preserving important image details. Techniques like spatial filtering, temporal noise reduction, and advanced denoising algorithms can be applied to reduce or minimize noise artifacts.

[0003] Conventionally, image denoising is done on single gray-scale images or demosaiced three-channel (such as red-green-blue or "RGB") images. However, many times, there is a need to denoise images while in a raw image format prior to demosaicing. Noise in raw images tends to be less correlated and therefore less challenging to handle. In addition, raw images have three times fewer unknown pixel values compared to RGB images. Nevertheless, what makes denoising images in raw format difficult is that these images include pixels of different colors adjacent to each other. The exact arrangement of different color channels on a pixel grid depends on the color filter array (CFA) used in or with an image sensor.

[0004] With the trend towards high mega-pixel cameras, such as 200 megapixel (MP) cameras, the latest image sensors are now adopting more novel color filter array patterns, such as the Quad Bayer (Tetra) CFA pattern or the Nona and Hexa Deca (Tetra2) CFA pattern. These new color filter array patterns offer the flexibility of binning adjacent pixels in the same color channel for better imaging signal in low light or short exposure scenarios.

[0005] Denoising these non-Bayer raw images presents unique challenges because of the very large number of pixels that need to be processed and because of the increased pixel distance between pixels of different colors. Many different types of denoising techniques have been proposed, but most of those employing deep neural networks (DNNs) are trained for RGB images or traditional Bayer pattern raw images. Traditionally, in order to train a deep neural network on Bayer raw images, the practice has been to separate individual color phases into four-channel (such as red-green-green-blue or "RGGB") images. However, a crucial shortcoming of such an approach is that separating out the sixteen channels in a Quad Bayer image results in aliasing artifacts due to spatial downsampling. FIGURES 17A and 17B and FIGURES 18A and 18B demonstrate example artifacts that can be caused due to noise in short exposure images. FIGURE 17A and FIGURE 18A are short exposure images, while FIGURE 17B and FIGURE 18B are corresponding long exposure images.

[0006] This disclosure relates to a detail preserving denoiser for short exposure images.

[0007] In an embodiment, a method may include obtaining a noisy training image and a ground truth image. The method may include converting the noisy training image into a first plurality of color channels. The method may include generating, using an artificial intelligence (AI)-based denoiser, denoised images from the first plurality of color channels, where each denoised image corresponds to a respective color channel of the first plurality of color channels. The method may include generating a noisy ground truth image from the ground truth image. The method may include converting the noisy ground truth image into a second plurality of color channels. The method may include determining a loss based on a comparison between the denoised images and the second plurality of color channels. The method may include adapting weights of the AI-based denoiser based on the loss determined based on the comparison between the denoised images and the second plurality of color channels.

[0008] In an embodiment, an electronic device may include at least one processing device configured to train an AI-based denoiser. To train the AI-based denoiser, the at least one processing device may be configured to obtain a noisy training image and a ground truth image. To train the AI-based denoiser, the at least one processing device may be configured to convert the noisy training image into a first plurality of color channels. To train the AI-based denoiser, the at least one processing device may be configured to generate, using the AI-based denoiser, denoised images from the first plurality of color channels, wherein each denoised image corresponds to a respective color channel of the first plurality of color channels. To train the AI-based denoiser, the at least one processing device may be configured to generate a noisy ground truth image from the ground truth image. To train the AI-based denoiser, the at least one processing device may be configured to convert the noisy ground truth image into a second plurality of color channels. To train the AI-based denoiser, the at least one processing device may be configured to determine a loss based on a comparison between the denoised images and the second plurality of color channels. To train the AI-based denoiser, the at least one processing device may be configured to adapt weights of the AI-based denoiser based on the loss determined based on the comparison between the denoised images and the second plurality of color channels.

[0009] In an embodiment, a computer-readable storage medium storing instructions is provided. The instructions, when executed by at least one processor, may cause the at least one processor to perform the method corresponding.

[0010] Other technical features may be readily apparent to one skilled in the art from the following figures, descriptions, and claims.

[0011] For a more complete understanding of this disclosure and its advantages, reference is now made to the following description taken in conjunction with the accompanying drawings, in which like reference numerals represent like parts:

[0012] FIGURE 1 illustrates an example network configuration providing a detail-preserving denoiser for short exposure images processing in accordance with this disclosure;

[0013] FIGURE 2 illustrates an example process of training a detail-preserving artificial intelligence / machine learning (AI / ML)-based denoising model for short exposure images in accordance with this disclosure;

[0014] FIGURE 3 illustrates an example process of using a detail-preserving AI / ML-based denoising model in accordance with this disclosure;

[0015] FIGURE 4 illustrates an example framework for training a detail-preserving AI / ML-based denoising model in accordance with this disclosure;

[0016] FIGURE 5 illustrates an example architecture for employing a detail-preserving AI / ML-based denoising model in accordance with this disclosure;

[0017] FIGURE 6 illustrates an example denoise network for detail-preserving AI / ML-based denoising in accordance with this disclosure;

[0018] FIGURE 7 illustrates an example grouping of Tetra pixels for detail-preserving AI / ML-based denoising in accordance with this disclosure;

[0019] FIGURE 8 illustrates an example of using a neighborhood of pixels for a corresponding pixel in a ground truth image to compute inter-total variance loss in accordance with this disclosure;

[0020] FIGURES 9A and 9B illustrate example pixel groupings for detail-preserving AI / ML-based denoising of CFA patterns other than Tetra in accordance with this disclosure;

[0021] FIGURES 10A and 10B illustrate an example addition of Gaussian noise with zero mean and a given standard deviation to a ground truth image in accordance with this disclosure;

[0022] FIGURES 11A and 11B illustrate an example detail-preserving denoising for short exposure images in accordance with this disclosure;

[0023] FIGURES 12A and 12B illustrate an example performance of detail-preserving denoising for short exposure images in accordance with this disclosure;

[0024] FIGURES 13A and 13B illustrate an example performance of detail-preserving denoising with noise regularization in accordance with this disclosure;

[0025] FIGURES 14A and 14B illustrate an example use of a Bayer denoising pipeline to denoise Tetra images;

[0026] FIGURES 15A and 15B illustrate an example of using a 4Х4 Tetra kernel pattern as a single-channel input and separating each color out as a channel input;

[0027] FIGURE 16A illustrates an example image (and an enlarged portion thereof) denoised using a Bayer denoiser by splitting the Tetra image into four Bayer images;

[0028] FIGURE 16B illustrates an example image (and an enlarged portion thereof) denoised by first separating each color into a separate channel; and

[0029] FIGURES 17A and 17B and FIGURES 18A and 18B illustrate example artifacts that can be caused due to noise in short exposure images.

[0030] Before undertaking the DETAILED DESCRIPTION below, it may be advantageous to set forth definitions of certain words and phrases used throughout this patent document. The terms "transmit," "receive," and "communicate," as well as derivatives thereof, encompass both direct and indirect communication. The terms "include" and "comprise," as well as derivatives thereof, mean inclusion without limitation. The term "or" is inclusive, meaning and / or. The phrase "associated with," as well as derivatives thereof, means to include, be included within, interconnect with, contain, be contained within, connect to or with, couple to or with, be communicable with, cooperate with, interleave, juxtapose, be proximate to, be bound to or with, have, have a property of, have a relationship to or with, or the like.

[0031] Moreover, various functions described below can be implemented or supported by one or more computer programs, each of which is formed from computer readable program code and embodied in a computer readable medium. The terms "application" and "program" refer to one or more computer programs, software components, sets of instructions, procedures, functions, objects, classes, instances, related data, or a portion thereof adapted for implementation in a suitable computer readable program code. The phrase "computer readable program code" includes any type of computer code, including source code, object code, and executable code. The phrase "computer readable medium" includes any type of medium capable of being accessed by a computer, such as read only memory (ROM), random access memory (RAM), a hard disk drive, a compact disc (CD), a digital video disc (DVD), or any other type of memory. A "non-transitory" computer readable medium excludes wired, wireless, optical, or other communication links that transport transitory electrical or other signals. A non-transitory computer readable medium includes media where data can be permanently stored and media where data can be stored and later overwritten, such as a rewritable optical disc or an erasable memory device.

[0032] As used here, terms and phrases such as "have," "may have," "include," or "may include" a feature (like a number, function, operation, or component such as a part) indicate the existence of the feature and do not exclude the existence of other features. Also, as used here, the phrases "A or B," "at least one of A and / or B," or "one or more of A and / or B" may include all possible combinations of A and B. For example, "A or B," "at least one of A and B," and "at least one of A or B" may indicate all of (1) including at least one A, (2) including at least one B, or (3) including at least one A and at least one B. Further, as used here, the terms "first" and "second" may modify various components regardless of importance and do not limit the components. These terms are only used to distinguish one component from another. For example, a first user device and a second user device may indicate different user devices from each other, regardless of the order or importance of the devices. A first component may be denoted a second component and vice versa without departing from the scope of this disclosure.

[0033] It will be understood that, when an element (such as a first element) is referred to as being (operatively or communicatively) "coupled with / to" or "connected with / to" another element (such as a second element), it can be coupled or connected with / to the other element directly or via a third element. In contrast, it will be understood that, when an element (such as a first element) is referred to as being "directly coupled with / to" or "directly connected with / to" another element (such as a second element), no other element (such as a third element) intervenes between the element and the other element.

[0034] As used here, the phrase "configured (or set) to" may be interchangeably used with the phrases "suitable for," "having the capacity to," "designed to," "adapted to," "made to," or "capable of" depending on the circumstances. The phrase "configured (or set) to" does not essentially mean "specifically designed in hardware to." Rather, the phrase "configured to" may mean that a device can perform an operation together with another device or parts. For example, the phrase "processor configured (or set) to perform A, B, and C" may mean a generic-purpose processor (such as a CPU or application processor) that may perform the operations by executing one or more software programs stored in a memory device or a dedicated processor (such as an embedded processor) for performing the operations.

[0035] The terms and phrases as used here are provided merely to describe some embodiments of this disclosure but not to limit the scope of other embodiments of this disclosure. It is to be understood that the singular forms "a," "an," and "the" include plural references unless the context clearly dictates otherwise. All terms and phrases, including technical and scientific terms and phrases, used here have the same meanings as commonly understood by one of ordinary skill in the art to which the embodiments of this disclosure belong. It will be further understood that terms and phrases, such as those defined in commonly used dictionaries, should be interpreted as having a meaning that is consistent with their meaning in the context of the relevant art and will not be interpreted in an idealized or overly formal sense unless expressly so defined here. In some cases, the terms and phrases defined here may be interpreted to exclude embodiments of this disclosure.

[0036] Examples of an "electronic device" according to embodiments of this disclosure may include at least one of a smartphone, a tablet personal computer (PC), a mobile phone, a video phone, an e-book reader, a desktop PC, a laptop computer, a netbook computer, a workstation, a personal digital assistant (PDA), a portable multimedia player (PMP), an MP3 player, a mobile medical device, a camera, or a wearable device (such as smart glasses, a head-mounted device (HMD), electronic clothes, an electronic bracelet, an electronic necklace, an electronic accessory, an electronic tattoo, a smart mirror, or a smart watch). Other examples of an electronic device include a smart home appliance. Examples of the smart home appliance may include at least one of a television, a digital video disc (DVD) player, an audio player, a refrigerator, an air conditioner, a cleaner, an oven, a microwave oven, a washer, a dryer, an air cleaner, a set-top box, a home automation control panel, a security control panel, a TV box (such as SAMSUNG HOMESYNC, APPLETV, or GOOGLE TV), a smart speaker or speaker with an integrated digital assistant (such as SAMSUNG GALAXY HOME, APPLE HOMEPOD, or AMAZON ECHO), a gaming console (such as an XBOX, PLAYSTATION, or NINTENDO), an electronic dictionary, an electronic key, a camcorder, or an electronic picture frame. Still other examples of an electronic device include at least one of various medical devices (such as diverse portable medical measuring devices (like a blood sugar measuring device, a heartbeat measuring device, or a body temperature measuring device), a magnetic resource angiography (MRA) device, a magnetic resource imaging (MRI) device, a computed tomography (CT) device, an imaging device, or an ultrasonic device), a navigation device, a global positioning system (GPS) receiver, an event data recorder (EDR), a flight data recorder (FDR), an automotive infotainment device, a sailing electronic device (such as a sailing navigation device or a gyro compass), avionics, security devices, vehicular head units, industrial or home robots, automatic teller machines (ATMs), point of sales (POS) devices, or Internet of Things (IoT) devices (such as a bulb, various sensors, electric or gas meter, sprinkler, fire alarm, thermostat, street light, toaster, fitness equipment, hot water tank, heater, or boiler). Other examples of an electronic device include at least one part of a piece of furniture or building / structure, an electronic board, an electronic signature receiving device, a projector, or various measurement devices (such as devices for measuring water, electricity, gas, or electromagnetic waves). Note that, according to various embodiments of this disclosure, an electronic device may be one or a combination of the above-listed devices. According to some embodiments of this disclosure, the electronic device may be a flexible electronic device. The electronic device disclosed here is not limited to the above-listed devices and may include new electronic devices depending on the development of technology.

[0037] In the following description, electronic devices are described with reference to the accompanying drawings, according to various embodiments of this disclosure. As used here, the term "user" may denote a human or another device (such as an artificial intelligent electronic device) using the electronic device.

[0038] Definitions for other certain words and phrases may be provided throughout this patent document. Those of ordinary skill in the art should understand that in many if not most instances, such definitions apply to prior as well as future uses of such defined words and phrases.

[0039] None of the description in this application should be read as implying that any particular element, step, or function is an essential element that must be included in the claim scope. The scope of patented subject matter is defined only by the claims. Moreover, none of the claims is intended to invoke 35 U.S.C. § 112(f) unless the exact words "means for" are followed by a participle. Use of any other term, including without limitation "mechanism," "module," "device," "unit," "component," "element," "member," "apparatus," "machine," "system," "processor," or "controller," within a claim is understood by the Applicant to refer to structures known to those skilled in the relevant art and is not intended to invoke 35 U.S.C. § 112(f).

[0040] The processor may include various processing circuitry and / or multiple processors. For example, as used herein, including the claims, the term "processor" may include various processing circuitry, including at least one processor, wherein one or more of at least one processor, individually and / or collectively in a distributed manner, may be configured to perform various functions described herein. As used herein, when "a processor", "at least one processor", and "one or more processors" are described as being configured to perform numerous functions, these terms cover situations, for example and without limitation, in which one processor performs some of recited functions and another processor(s) performs other of recited functions, and also situations in which a single processor may perform all recited functions. Additionally, the at least one processor may include a combination of processors performing various of the recited / disclosed functions, e.g., in a distributed manner. At least one processor may execute program instructions to achieve or perform various functions.

[0041] The figures, discussed below, and the various embodiments of this disclosure are described with reference to the accompanying drawings. However, it should be appreciated that this disclosure is not limited to these embodiments, and all changes and / or equivalents or replacements thereto also belong to the scope of this disclosure. The same or similar reference denotations may be used to refer to the same or similar elements throughout the specification and the drawings.

[0042] Currently, Quad Bayer sensors (also referred to as Tetra sensors in this disclosure) are gaining popularity. Tetra sensors offer the flexibility of extremely high-resolution captures (compared to, for example, Bayer sensors or other types of sensors) while also allowing for binning of pixels for low light conditions to increase the signal-to-noise ratio (SNR) by trading off resolution. Given the very high resolutions in Tetra sensors, captured image data can be noisier compared to typical (such as 12MP) Bayer sensors due to comparatively small pixel size.

[0043] FIGURES 14A and 14B illustrate an example use of a Bayer denoising pipeline to denoise Tetra images. More specifically, FIGURE 14A illustrates format conversions involved, and FIGURE 14B diagrammatically illustrates a pipeline. An example Tetra kernel pattern 1401 is illustrated on the left side of the format conversions 1400 depicted in FIGURE 14A and includes an arrangement of color filters on a 4Х4 pixel array. In a Tetra sensor, each 4Х4 pixel array 1401 may include multiple 2Х2 arrays 1402-1405. Each 2Х2 array 1402-1405 may be clustered with the same color filter as shown in FIGURE 14A. For example, each 4Х4 Tetra kernel pattern 1401 may include one 2Х2 array (such as 2Х2 array 1404) having blue color filters, another 2Х2 array (such as 2Х2 array 1403) having red color filters, and two other 2Х2 arrays (such as 2Х2 array 1402 and 2Х2 array 1405) having green color filters.

[0044] In order to use a Bayer denoising pipeline 1425 in FIGURE 14B, each noisy Tetra image 1426 (composed of a plurality of 4Х4 pixel arrays, including a pixel array for the 4Х4 Tetra kernel pattern 1401) may be subjected to a Tetra-to-Bayer converter 1406, thus resulting in four noisy Bayer images 1407-1410. The four noisy Bayer images 1407-1410 may be respectively provided to four Bayer denoisers 1417-1420, which produce four denoised Bayer images 1427-1430. The four denoised Bayer images 1427-1430 may be used by a Bayer-to-Tetra converter 1411 to generate a denoised Tetra image 1431.

[0045] A Bayer denoising pipeline as illustrated by FIGURES 14A and 14B can be used to denoise Tetra images by breaking each Tetra image into four separate Bayer images (as illustrated in FIGURES 14A and 14B). In addition, existing techniques for training AI / ML networks for camera sensors also include using the entire Tetra pattern image as a single-channel input (such as shown in FIGURE 15A) or separating out pixels in each Tetra pattern into separate channels (such as shown in FIGURE 15B).

[0046] Using single-channel input and output (as in approach 1500 in FIGURE 15A) can result in better image quality as opposed to using multiple channel inputs (for example, using four Bayer channels obtained from the single Tetra image). However, using a single channel input may result in high computational cost because of the large memory requirements and consequently larger number of arithmetic operations that are needed. Consider the results shown in FIGURES 16A and 16B. FIGURE 16A may be an image (and an enlarged portion thereof) denoised using a Bayer denoiser by splitting the Tetra image into four Bayer images as shown in FIGURE 14A and FIGURE 14B, while FIGURE 16B may be an image (and an enlarged portion thereof) denoised by first separating each color four separate channels as shown in FIGURE 7. However, using a Bayer denoiser as illustrated by FIGURES 14A and 14B can have low computational costs but can suffer from aliasing artifacts and therefore detail loss. FIGURE 16A may be the result of using a Bayer denoiser as illustrated in FIGURE 14A.

[0047] The present disclosure may provide an efficient, detail-preserving trained AI / ML-based denoiser for Tetra images or other images and a new and unique framework for training such a denoiser. Among other things, an optimal way of grouping pixels can be designed to achieve a desired tradeoff between image quality and inference speed / computation. Moreover, due to a large number of pixels per image, the trained AI / ML-based denoiser might ordinarily require a large amount of training data, which is hard to obtain. Therefore, a new inter-channel loss function may be presented that can effectively regularize the AI / ML-based denoiser, causing the AI / ML-based denoiser to reduce or avoid noisy artifacts. Additionally, in order to avoid over-fitting, noise may be added to ground truth images during training of the AI / ML-based denoiser.

[0048] FIGURE 1 illustrates an example network configuration 100 providing a detail-preserving denoiser for short exposure images processing in accordance with this disclosure. The embodiment of the network configuration 100 shown in FIGURE 1 is for illustration only. Other embodiments of the network configuration 100 could be used without departing from the scope of this disclosure.

[0049] According to an embodiment of this disclosure, an electronic device 101 may be included in the network configuration 100. The electronic device 101 can include at least one of a bus 110, a processor 120, a memory 130, an input / output (I / O) interface 150, a display 160, a communication interface 170, or a sensor 180. In an embodiment, the electronic device 101 may exclude at least one of these components or may add at least one other component. The bus 110 may include a circuit for connecting the components 120-180 with one another and for transferring communications (such as control messages and / or data) between the components.

[0050] The processor 120 may include one or more processing devices, such as one or more microprocessors, microcontrollers, digital signal processors (DSPs), application specific integrated circuits (ASICs), or field programmable gate arrays (FPGAs). In an embodiment, the processor 120 may include one or more of a central processing unit (CPU), an application processor (AP), a communication processor (CP), or a graphics processor unit (GPU). The processor 120 may be able to perform control on at least one of the other components of the electronic device 101 and / or perform an operation or data processing relating to communication or other functions. As described in more detail below, the processor 120 may perform various operations related to a detail-preserving AI / ML-based denoiser for short exposure images.

[0051] The memory 130 can include a volatile and / or non-volatile memory. For example, the memory 130 can store commands or data related to at least one other component of the electronic device 101. According to embodiments of this disclosure, the memory 130 can store software and / or a program 140. The program 140 may include, for example, a kernel 141, middleware 143, an application programming interface (API) 145, and / or an application program (or "application") 147. At least a portion of the kernel 141, middleware 143, or API 145 may be denoted an operating system (OS).

[0052] The kernel 141 can control or manage system resources (such as the bus 110, processor 120, or memory 130) used to perform operations or functions implemented in other programs (such as the middleware 143, API 145, or application 147). The kernel 141 may provide an interface that allows the middleware 143, the API 145, or the application 147 to access the individual components of the electronic device 101 to control or manage the system resources. The application 147 may support various functions related to a detail-preserving AI / ML-based denoiser for short exposure images. These functions can be performed by a single application or by multiple applications that each carries out one or more of these functions. The middleware 143 can function as a relay to allow the API 145 or the application 147 to communicate data with the kernel 141, for instance. A plurality of applications 147 can be provided. The middleware 143 is able to control work requests received from the applications 147, such as by allocating the priority of using the system resources of the electronic device 101 (like the bus 110, the processor 120, or the memory 130) to at least one of the plurality of applications 147. The API 145 may be an interface allowing the application 147 to control functions provided from the kernel 141 or the middleware 143. For example, the API 145 may include at least one interface or function (such as a command) for filing control, window control, image processing, or text control.

[0053] The I / O interface 150 may serve as an interface that can, for example, transfer commands or data input from a user or other external devices to other component(s) of the electronic device 101. The I / O interface 150 can also output commands or data received from other component(s) of the electronic device 101 to the user or the other external device.

[0054] The display 160 may include, for example, a liquid crystal display (LCD), a light emitting diode (LED) display, an organic light emitting diode (OLED) display, a quantum-dot light emitting diode (QLED) display, a microelectromechanical systems (MEMS) display, or an electronic paper display. The display 160 can also be a depth-aware display, such as a multi-focal display. The display 160 is able to display, for example, various contents (such as text, images, videos, icons, or symbols) to the user. The display 160 can include a touchscreen and may receive, for example, a touch, gesture, proximity, or hovering input using an electronic pen or a body portion of the user.

[0055] The communication interface 170, for example, is able to set up communication between the electronic device 101 and an external electronic device (such as a first electronic device 102, a second electronic device 104, or a server 106). For example, the communication interface 170 can be connected with a network 162 or 164 through wireless or wired communication to communicate with the external electronic device. The communication interface 170 can be a wired or wireless transceiver or any other component for transmitting and receiving signals.

[0056] The wireless communication is able to use at least one of, for example, WiFi, long term evolution (LTE), long term evolution-advanced (LTE-A), 5th generation wireless system (5G), millimeter-wave or 60 GHz wireless communication, Wireless USB, code division multiple access (CDMA), wideband code division multiple access (WCDMA), universal mobile telecommunication system (UMTS), wireless broadband (WiBro), or global system for mobile communication (GSM), as a communication protocol. The wired connection can include, for example, at least one of a universal serial bus (USB), high definition multimedia interface (HDMI), recommended standard 232 (RS-232), or plain old telephone service (POTS). The network 162 or 164 may include at least one communication network, such as a computer network (like a local area network (LAN) or wide area network (WAN)), Internet, or a telephone network.

[0057] The electronic device 101 may include one or more sensors 180 that can meter a physical quantity or detect an activation state of the electronic device 101 and convert metered or detected information into an electrical signal. For example, one or more sensors 180 can include one or more cameras or other imaging sensors for capturing images of scenes. The sensor(s) 180 can also include one or more buttons for touch input, one or more microphones, a gesture sensor, a gyroscope or gyro sensor, an air pressure sensor, a magnetic sensor or magnetometer, an acceleration sensor or accelerometer, a grip sensor, a proximity sensor, a color sensor (such as an RGB sensor), a bio-physical sensor, a temperature sensor, a humidity sensor, an illumination sensor, an ultraviolet (UV) sensor, an electromyography (EMG) sensor, an electroencephalogram (EEG) sensor, an electrocardiogram (ECG) sensor, an infrared (IR) sensor, an ultrasound sensor, an iris sensor, or a fingerprint sensor. The sensor(s) 180 can include an inertial measurement unit, which can include one or more accelerometers, gyroscopes, and other components. In addition, the sensor(s) 180 can include a control circuit for controlling at least one of the sensors included here. Any of these sensor(s) 180 can be located within the electronic device 101.

[0058] In an embodiment, the first external electronic device 102 or the second external electronic device 104 can be a wearable device or an electronic device-mountable wearable device (such as a head mounted display (or "HMD")). When the electronic device 101 is mounted in the electronic device 102 (such as the HMD), the electronic device 101 can communicate with the electronic device 102 through the communication interface 170. The electronic device 101 can be directly connected with the electronic device 102 to communicate with the electronic device 102 without involving with a separate network. The electronic device 101 can also be an augmented reality wearable device, such as eyeglasses, which include one or more imaging sensors, or a VR or XR headset.

[0059] The first and second external electronic devices 102 and 104 and the server 106 each can be a device of the same or a different type from the electronic device 101. According to an embodiment of this disclosure, the server 106 may include a group of one or more servers. Also, according to an embodiment of this disclosure, all or some of the operations executed on the electronic device 101 can be executed on another or multiple other electronic devices (such as the electronic devices 102 and 104 or server 106). Further, according to an embodiment of this disclosure, when the electronic device 101 should perform some function or service automatically or at a request, the electronic device 101, instead of executing the function or service on its own or additionally, can request another device (such as electronic devices 102 and 104 or server 106) to perform at least some functions associated therewith. The other electronic device (such as electronic devices 102 and 104 or server 106) is able to execute the requested functions or additional functions and transfer a result of the execution to the electronic device 101. The electronic device 101 can provide a requested function or service by processing the received result as it is or additionally. To that end, a cloud computing, distributed computing, or client-server computing technique may be used, for example. While FIGURE 1 shows that the electronic device 101 includes the communication interface 170 to communicate with the external electronic devices 102 and 104 or server 106 via the network 162 or 164, the electronic device 101 may be independently operated without a separate communication function according to an embodiment of this disclosure.

[0060] The server 106 can include the same or similar components 110-180 as the electronic device 101 (or a suitable subset thereof). The server 106 can support the electronic device 101 by performing at least one of operations (or functions) implemented on the electronic device 101. For example, the server 106 can include a processing module or processor that may support the processor 120 implemented in the electronic device 101. As described in more detail below, the server 106 may perform various operations related to a detail-preserving AI / ML-based denoiser for short exposure images.

[0061] Although FIGURE 1 illustrates one example of a network configuration 100 including an electronic device 101, various changes may be made to FIGURE 1. For example, the network configuration 100 could include any number of each component in any suitable arrangement. In general, computing and communication systems come in a wide variety of configurations, and FIGURE 1 does not limit the scope of this disclosure to any particular configuration. Also, while FIGURE 1 illustrates one operational environment in which various features disclosed in this patent document can be used, these features could be used in any other suitable system.

[0062] FIGURE 2 illustrates an example process 200 of training a detail-preserving AI / ML-based denoising model for short exposure images in accordance with this disclosure. For ease of explanation, the process 200 of FIGURE 2 is described as being performed using the server 106 in the network configuration 100 of FIGURE 1. However, the process 200 may be performed using any other suitable device(s) (such as the electronic device 101) and in any other suitable system(s).

[0063] As shown in FIGURE 2, the process 200 may begin with obtaining a noisy training image and a corresponding ground truth image (operation 201). The noisy training image and the ground truth image may have an identical color filter array (CFA) pattern, such as a Tetra pattern, a Hexa-Deca pattern, or a Nona pattern. The noisy training image may be an exposure value zero (EV0) image or a lower exposure value (e.g., EV-2) image. The noisy training image may be one of a plurality of noisy training images, each noisy training image having a different proportion of noise. The noisy training image may be converted into a first plurality of color channels (operation 202). For example, a noisy Tetra image may be converted into four separate color channels. Note that the number of color channels here can vary based on the images being processed.

[0064] Using an AI / ML-based denoiser, denoised images may be generated from the first plurality of color channels (operation 203). Each denoised image corresponds to a respective color channel of the first plurality of color channels. A noisy ground truth image may be generated from the ground truth image (operation 204). For example, the noisy ground truth image may be generated by adding zero-mean Gaussian noise to the ground truth image. The noisy ground truth image may be converted in a second plurality of color channels (operation 205). For instance, if the noisy training image is converted into four color channels, the noisy ground truth image can also be converted into the four corresponding color channels. Again, note that the number of color channels here can vary based on the images being processed.

[0065] A loss may be determined based on a comparison between the denoised images and the second plurality of color channels into which the noisy ground truth image was converted (operation 206). For example, the loss may be determined using a linear combination of a mean absolute error (L1) loss, a multi-scale structural similarity loss, and an inter-channel loss, which can constitute a total variation loss between each of the denoised images and a corresponding color channel for the ground truth image. Weights of the AI / ML-based denoiser may be adapted based on the loss determined by the comparison of the denoised images and the second plurality of color channels (operation 207). In an embodiment, noise-based regularization to control overfitting to the training data can be used to produce superior weight adaptation.

[0066] Although FIGURE 2 illustrates one example of a process 200 of training a detail-preserving AI / ML-based denoising model, various changes may be made to FIGURE 2. For example, while shown as a series of operations, various operations in FIGURE 2 could overlap, occur in parallel, occur in a different order, or occur any number of times (including zero times).

[0067] FIGURE 3 illustrates an example process 300 of using a detail-preserving AI / ML-based denoising model in accordance with this disclosure. For ease of explanation, the process 300 of FIGURE 3 is described as being performed using the electronic device 101 in the network configuration 100 of FIGURE 1. For example, the process 300 may be performed by the electronic device 101 using an AI / ML model that is trained using the process 200 of FIGURE 2. However, the process 300 may be performed using any other suitable device(s) and in any other suitable system(s).

[0068] As shown in FIGURE 3, the process 300 may begin with obtaining a noisy captured image (operation 301). For example, a short exposure image or other image may be captured using one or more imaging sensors 180 of the electronic device 101. The noisy captured image may be converted from a CFA pattern into a plurality of color channels (operation 302). For example, the noisy captured image may be converted into four color channels. Examples of suitable color channel pixel groupings are depicted and described in connection with FIGURES 7 and 9A-9B. In an embodiment, this operation may be performed for a subblock of pixels within the noisy captured image, such as a 4Х4 block of pixels for a Tetra CFA pattern, an 8Х8 block of pixels for a Hexa-Deca CFA, or a 6Х6 block of pixels for a Nona CFA pattern.

[0069] The plurality of color channels from the noisy captured image may be provided to an AI / ML-based denoiser that was trained using the process 200 of FIGURE 2 or a similar process (operation 303). For example, the AI / ML-based denoiser may be trained using separate color channels and one or more of ground truth or noisy ground truth. In an embodiment, the AI / ML-based denoiser may be trained using a loss function that is a linear combination of a mean absolute error (L1) loss, a multi-scale structural similarity loss, and an inter-channel loss. The plurality of color channels provided to the AI / ML-based decoder can be denoised by the AI / ML-based denoiser (operation 304). For example, the four color channels may be denoised using the AI / ML-based denoiser. In an embodiment, the denoised color channel(s) can represent a prediction (or inference) by the AI / ML based denoiser based on an inter-total variance loss that takes into account neighborhood pixels of other color channels within the captured noisy image. The denoised color channels may be converted into a denoised image (operation 305).

[0070] Although FIGURE 3 illustrates one example of a process 300 of using a detail-preserving AI / ML-based denoising model, various changes may be made to FIGURE 3. For example, while shown as a series of operations, various operations in FIGURE 3 could overlap, occur in parallel, occur in a different order, or occur any number of times (including zero times).

[0071] FIGURE 4 illustrates an example framework 400 for training a detail-preserving AI / ML-based denoising model in accordance with this disclosure, and FIGURE 5 illustrates an example architecture 500 for employing a detail-preserving AI / ML-based denoising model in accordance with this disclosure. For ease of explanation, the framework 400 of FIGURE 4 is described as being implemented using the server 106 in the network configuration 100 of FIGURE 1, and the architecture 500 of FIGURE 5 is described as being implemented using the electronic device 101 in the network configuration 100 of FIGURE 1. However, the framework 400 and architecture 500 may be implemented using any other suitable device(s) and in any other suitable system(s).

[0072] As shown in FIGURE 4, this example training of a detail-preserving AI / ML-based denoising model may use, as inputs, a noisy Tetra image 401 and a corresponding ground truth Tetra image 402. The noisy Tetra image 401 may be converted to a four-channel image using a converter 403. In an embodiment, the converter 403 may implement a new technique to group Tetra pixels for deep neural network processing, where pixels of the same color space are grouped as a single channel (representing a spatial resolution trade-off with depth). Such grouping may provide a good tradeoff between image quality and processing time. An example of the grouping of Tetra pixels that may be implemented by the converter 403 (both instances depicted in FIGURE 4) is described below in connection with FIGURE 7. The four-channel output of the converter 403 may be provided to a denoise network 404. In an embodiment, the denoise network 404 may be a deep neural network (DNN) or other AI / ML model, such as one having the structure and operation described below in connection with FIGURE 6. The denoise network 404 may produce a denoised output 405.

[0073] The framework 400 may also employ a new inter-channel loss function 406 to preserve better details while controlling noise. Instead of using total variation loss over the predicted output, total variation loss between the predicted denoised output 405 and the ground truth Tetra image 402 may be utilized, which has a regularizing effect. Also, in an embodiment, noise-based regularization 407 can be employed to control overfitting to the training data. Robust regularization may be advantageous given the limited amount of training data. Accordingly, noise may be added to the ground truth Tetra image 402 by the noise-based regularization 407. The noise-based regularization 407 may, for example, add zero-mean Gaussian noise to the ground truth Tetra image 402. Adding noise to the ground truth Tetra image 402 may have the effect of increasing the variance in the data and the generalizability of the resulting model. The loss function 406 may compare corresponding color channels for the ground truth image and the denoised image in determining total variation loss. In an embodiment, the loss function 406 may implement a linear combination of a mean absolute error (L1) loss, a multi-scale structural similarity loss, and an inter-channel loss.

[0074] Note that multiple training data pairs may be used to train the denoise network 404. For example, both the noisy Tetra image 401 (such as EV-2 IN) and the ground truth Tetra image 402 (such as EV0 GT) may form one training data pair, and the output of the noise-based regularization 407 (such as EV0 IN) and the ground truth Tetra image 402 (such as EV0 GT) may form another training data pair. This approach may help to ensure that the AI / ML-based denoiser produced by the framework 400 is able to handle both higher noise and medium noise inputs since the denoiser has been trained to map each of these inputs to a ground truth image. The ground truth Tetra image 402 and the output of the noise-based regularization 407 may be converted to the same four-channel pixel groupings as the output of the converters 403 in FIGURE 4 for use by both the loss function 406 and training of the denoise network.

[0075] As shown in FIGURE 5, the architecture 500 may receive a noisy Tetra image 501 to be denoised. The received noisy Tetra image 501 may be processed by the converter 403, which converts the noisy Tetra image 501 into four channels or groupings of pixels of the same color space. The groupings of pixels output by the converter 403 may be provided to a denoiser 502, which represents the trained denoise network 404. The output of the denoiser 502 may be supplied to a four channel-to-Tetra converter 503, which implements the converse operation of the converter 403. The output of the converter 503 may be a denoised Tetra image 504.

[0076] Although FIGURE 4 illustrates one example of a framework 400 for training a detail-preserving AI / ML-based denoising model and FIGURE 5 illustrates one example of an architecture 500 for employing a detail-preserving AI / ML-based denoising model, various changes may be made to FIGURES 4 and 5. For example, while the examples of FIGURES 4 and 5 are described in the context of Tetra image CFA patterns, the same framework 400 and architecture 500 may be employed with other CFA patterns. For instance, the framework 400 and the architecture 500 may be implemented for a Hexa-Deca CFA pattern or a Nona CFA pattern. In an embodiment, identical CFA patterns may be used for both the noisy image and the ground truth image.

[0077] FIGURE 6 illustrates an example denoise network 600 for detail-preserving AI / ML-based denoising in accordance with this disclosure. In an embodiment, the denoise network 600 shown in FIGURE 6 may be employed as the denoise network 404 within the framework 400 of FIGURE 4 and as the denoiser 502 within the architecture 500 of FIGURE 5.

[0078] As shown in FIGURE 6, this example of the denoise network 600 may have a U-net based architecture with skip connections. The denoise network 600 here may include an encoder 601 (or contracting path) and a decoder 602 (or expanding path). The encoder 601 may include convolutional layers and strided convolutional layers. Each convolutional layer may apply a convolution operation to its input (such as a portion of an image or subarray of pixels from a color channel) and pass the result to the next layer. Each strided convolutional layer may use a filter with a step size (stride) corresponding to a number of pixels skipped in the convolution operation. The decoder 602 may include transpose convolutional layers (also referred to as "deconvolutional" or "up-convolutional" layers) and concatenation functions with skip connections. Transpose convolutions may up-sample image data to recover spatial information, and skip connections are used to send image data directly from the contracting path to a corresponding level in the expanding path without passing through all intervening layers, allowing features to be preserved and avoid information loss as a result of the contracting path. For example, a feature map from a convolutional operation can be concatenated with image data from a skip connection.

[0079] In the example of FIGURE 6, the encoder 601 for the denoise network 600 may receive a noisy image and include a first convolutional layer 603 paired with a first strided convolutional layer 604, a second convolutional layer 605 paired with a second strided convolutional layer 606, a third convolutional layer 607 paired with a third strided convolutional layer 608, a fourth convolutional layer 609 paired with a fourth strided convolutional layer 610, and a fifth convolutional layer 611 paired with a fifth strided convolutional layer 612. Each strided convolutional layer 604, 606, 608, 610, 612 may use any suitable stride.

[0080] The decoder 602 for the denoise network 600 in the example of FIGURE 6 may include a fifth transpose convolutional layer 613, the output of which may be concatenated 614 with the skip connection image data from the fourth strided convolutional layer 610 and provided as an input to a fifth convolutional layer 615. The output of the fifth convolutional layer 615 may be received by a fourth transpose convolutional layer 616, the output of which may be concatenated 617 with the skip connection image data from the third strided convolutional layer 608 and provided as an input to a fourth convolutional layer 618. The output of the fourth convolutional layer 618 may be received by a third transpose convolutional layer 619, the output of which may be concatenated 620 with the skip connection image data from the second strided convolutional layer 606 and the third convolutional layer 607, and provided as an input to a third convolutional layer 621. The output of the third convolutional layer 621 may be received by a second transpose convolutional layer 622, the output of which may be concatenated 623 with the skip connection image data from the first strided convolutional layer 604 and provided as an input to a second convolutional layer 624. The output of the second convolutional layer 624 may be received by a first transpose convolutional layer 625, the output of which may be concatenated 626 with the skip connection image data from the first convolutional layer 603 and provided as an input to a first convolutional layer 627. The output of the first convolutional layer 627 may be a prediction of a denoised image.

[0081] In this disclosure, the denoise network 600 may operate sequentially on each color channel of a noisy Tetra image or other image for which pixel information has been grouped using a Tetra-to-four channel converter 403 or other converter. The output from each iteration of operation of the denoise network 600 can represent a single color channel of the denoised Tetra image or other denoised image. The complete denoised Tetra image or other complete denoised image can be formed by the converse pixel grouping of the four channel-to-Tetra converter 503 or other converter. In other cases, separate instances of the denoise network 600 may be implemented for each of the color channels, and the separate instances of the denoise network 600 may operate in parallel.

[0082] Although FIGURE 6 illustrates one example of a denoise network 600 for detail-preserving AI / ML-based denoising, various changes may be made to FIGURE 6. For example, a denoise network may implement any other suitable AI / ML-based architecture. Moreover, the depth, width, and sizes of different filters or layers can have any suitable values, such as those determined experimentally and chosen according to computational budgets.

[0083] FIGURE 7 illustrates an example grouping of Tetra pixels for detail-preserving AI / ML-based denoising in accordance with this disclosure. In an embodiment, the pixel grouping 700 may form the basis of operation for the converter(s) 403 within the framework 400 of FIGURE 4 and for the converter 403 and converter 503 within the architecture 500 of FIGURE 5. Also, in an embodiment, the pixel grouping 700 may be employed by the denoise network 600 of FIGURE 6.

[0084] As shown in FIGURE 7, the proposed Tetra-to-four channel converter(s) 403 may group channels based on colors as input to the denoise network 600. The pixel grouping 700 may represent a new technique for grouping Tetra pixels for deep neural network processing or other processing. Here, for example, the left side of FIGURE 7 may be treated as input, and the right side of FIGURE 7 may be treated as output for the converter(s) 403. The converse operation can be performed according to the groupings shown in FIGURE 7 by the converter 503. The Tetra CFA pattern (illustrated on the left side of FIGURE 14A) - namely four colorsGr,R,B, andGb- are sensed in a 4Х4 array of pixels. In the pixel grouping 700, an 8Х8 grid of pixels from the Tetra CFA pattern may be treated as a single Tetra cell (the left side of FIGURE 7). The pixel grouping 700 can be used to generate four channel patterns (Gr,R,B, andGb), each an aggregated 4Х4 grid of a given color (the right side of FIGURE 7). As discussed above, grouping pixels of the same color as a single channel (trading off spatial resolution with depth) may provide a good balance between image quality and processing time.

[0085] Although FIGURE 7 illustrates one example of a grouping of Tetra pixels for detail-preserving AI / ML-based denoising, various changes may be made to FIGURE 7. For example, while the use of a Tetra CFA pattern is shown here, other CFA patterns may be used.

[0086] Referring back to FIGURE 4, in an embodiment, the loss function 406 may be based on a new inter-channel loss function to preserve better details while controlling noise. Stated differently, the loss function 406 may be implemented as an inter-channel total variance loss function to improve image details while suppressing artifacts. In an embodiment, the loss function 406 may be expressed as follows.

[0087]

[0088] Here, the term combines three terms (L1 loss , multi-scale structural similarity (MS-SSIM) loss , and inter-channel loss ) with three associated weights ( ). Once a loss is determined using the loss function , regular backpropagation or other technique may be used for the training process. The inter-channel loss is discussed in further detail below.

[0089] In one approach, given a pair of input and ground truth images, the input image is fed into a network, and the loss between the output image and the ground truth image are determined. Backpropagation or other technique is used to adjust the weights of the network based on the loss. Note that the loss here can be determined using any number of pairs of input and ground truth images. Therefore, the loss function in this approach is a combination of L1 norm and MS-SSIM losses. To improve image details, the inter-total variance loss function may also be used. The loss can be computed between the output and ground truth images on each color channel. For example, the loss may be computed, then backpropagation may be used to learn the weights. The weights of the AI / ML-based denoiser may be adapted based on the loss.

[0090] Based on total variance loss, it may be assumed that neighborhood pixels are supposed to be similar. However, the loss function in the approach above suffers from a smoothing effect. In order to overcome this, inter-total variance loss may use a neighborhood of pixels of each corresponding pixel in the ground truth image, which could be expressed in the following manner.

[0091]

[0092] Here, represents the number of pixels for each channel ( in the example of FIGURE 7), represents the channel index for one of the channels (Gr,R,B, andGbin the example of FIGURE 7), represents the output of the denoise network 600, represents the ground truth for the same channel as considered for , represents the index of the four-pixel group, and represent the pixel index within the four-pixel group. FIGURE 8 illustrates an example of using a neighborhood of pixels for a corresponding pixel in a ground truth image to compute inter-total variance loss in accordance with this disclosure.

[0093] It is noted that although FIGURE 7 depicts an example grouping of Tetra pixels for detail-preserving AI / ML-based denoising and that the discussion above of the loss function 406 relates to a Tetra denoiser, a similar approach can be used for other kernel patterns. For example, FIGURES 9A and 9B illustrate example pixel groupings for detail-preserving AI / ML-based denoising of CFA patterns other than Tetra in accordance with this disclosure. Other aspects shown in FIGURES 4 and 5 may be similar for other kernel patterns. In FIGURE 9A, a 8Х8 Hexa-Deca CFA pattern is converted to four channels of 4Х4 pixel groupings. In FIGURE 9B, a 6Х6 Nona CFA pattern is converted to four channels of 3Х3 pixel groupings. Note, however, that these CFA patterns are examples only and that other CFA patterns may be used.

[0094] Referring back to FIGURE 4, the noise-based regularization 407 can be used to control overfitting to the training data, and the framework 400 may add noise to the ground truth Tetra image 402 to regularize the trained model and improve details in denoised output image. FIGURES 10A and 10B illustrate an example addition of Gaussian noise with zero mean and a given standard deviation in accordance with this disclosure.

[0095] FIGURES 11A and 11B illustrate an example detail-preserving denoising for short exposure images in accordance with this disclosure. FIGURE 11A may be a noisy image, while FIGURE 11B may be the resultant denoised image that is produced using a denoiser trained according to the present disclosure. FIGURES 12A and 12B illustrate an example performance of detail-preserving denoising for short exposure images in accordance with this disclosure. FIGURE 12A may be a noisy image denoised by a regular Bayer denoiser, while FIGURE 12B may be the same noisy image denoised image that is produced using a Tetra denoiser trained according to the present disclosure. FIGURES 13A and 13B illustrate an example performance of detail-preserving denoising with noise regularization in accordance with this disclosure. FIGURE 13A may be a noisy image denoised using a Tetra denoiser trained without noise regularization, while FIGURE 13B may be the same noisy image denoised using a Tetra denoiser trained with noise regularization according to the present disclosure.

[0096] Although FIGURES 11A through 13B illustrate examples of detail-preserving denoising and related performance, various changes may be made to FIGURES 11A through 13B. For example, images of scenes can vary widely, and FIGURES 11A through 13B do not limit the scope of this disclosure to any particular denoising or performance characteristics.

[0097] It should be noted that the functions shown in the figures or described above can be implemented in an electronic device 101, 102, 104, server 106, or other device(s) in any suitable manner. For example, in an embodiment, at least some of the functions shown in the figures or described above can be implemented or supported using one or more software applications or other software instructions that are executed by the processor 120 of the electronic device 101, 102, 104, server 106, or other device(s). In an embodiment, at least some of the functions shown in the figures or described above can be implemented or supported using dedicated hardware components. In general, the functions shown in the figures or described above can be performed using any suitable hardware or any suitable combination of hardware and software / firmware instructions. Also, the functions shown in the figures or described above can be performed by a single device or by multiple devices.

[0098] In an embodiment, the method, wherein the noisy training image and the ground truth image may have an identical image color filter array (CFA) pattern. The method, wherein the identical image CFA pattern may include a Tetra CFA pattern, a Hexa-Deca CFA pattern, or a Nona CFA pattern.

[0099] In an embodiment, the method, wherein determining the loss may include using a linear combination of a mean absolute error (L1) loss, a multi-scale structural similarity loss, and an inter-channel loss.

[0100] In an embodiment, the method, wherein generating the noisy ground truth image may include adding zero-mean Gaussian noise to the ground truth image.

[0101] In an embodiment, the method, wherein the noisy training image may include an exposure value zero (EV0) image or a lower exposure value image.

[0102] In an embodiment, the method, wherein determining the loss may include using a total variation loss between the denoised images and a corresponding color channel for the ground truth image.

[0103] In an embodiment, the method, wherein the noisy training image may include one of a plurality of noisy training images, each noisy training image having a different proportion of noise.

[0104] In an embodiment, the electronic device, wherein the noisy training image and the ground truth image may have an identical image color filter array (CFA) pattern. The electronic device, wherein the identical image CFA pattern may include a Tetra CFA pattern, a Hexa-Deca CFA pattern, or a Nona CFA pattern.

[0105] In an embodiment, the electronic device, wherein to determine the loss, the at least one processing device may be configured to use a linear combination of a mean absolute error (L1) loss, a multi-scale structural similarity loss, and an inter-channel loss.

[0106] In an embodiment, the electronic device, wherein, to generate the noisy ground truth image, the at least one processing device may be configured to add zero-mean Gaussian noise to the ground truth image.

[0107] In an embodiment, the electronic device, wherein the noisy training image may include an exposure value zero (EV0) image or a lower exposure value image.

[0108] In an embodiment, the electronic device, to determine the loss, the at least one processing device is configured to use a total variation loss between the denoised images and a corresponding color channel for the ground truth image.

[0109] In an embodiment, the electronic device, wherein the noisy training image may include one of a plurality of noisy training images, each noisy training image having a different proportion of noise.

[0110] In an embodiment, a method may include obtaining a noisy captured image. The method may include denoising the noisy captured image using an AI-based denoiser. The AI-based denoiser may be trained by obtaining a noisy training image and a ground truth image. The AI-based denoiser may be trained by converting the noisy training image into a first plurality of color channels. The AI-based denoiser may be trained by generating, using the AI-based denoiser, denoised images from the first plurality of color channels, where each denoised image corresponds to a respective color channel of the first plurality of color channels. The AI-based denoiser may be trained by generating a noisy ground truth image from the ground truth image. The AI-based denoiser may be trained by converting the noisy ground truth image into a second plurality of color channels. The AI-based denoiser may be trained by determining a loss based on a comparison between the denoised images and the second plurality of color channels. The AI-based denoiser may be trained by adapting weights of the AI-based denoiser based on the loss determined based on the comparison between the denoised images and the second plurality of color channels.

[0111] Although this disclosure has been described with reference to various example embodiments, various changes and modifications may be suggested to one skilled in the art. It is intended that this disclosure encompass such changes and modifications as fall within the scope of the appended claims.

Claims

1.A method comprising:obtaining a noisy training image and a ground truth image;converting the noisy training image into a first plurality of color channels;generating, using an artificial intelligence (AI)-based denoiser, denoised images from the first plurality of color channels, wherein each denoised image corresponds to a respective color channel of the first plurality of color channels;generating a noisy ground truth image from the ground truth image;converting the noisy ground truth image into a second plurality of color channels;determining a loss based on a comparison between the denoised images and the second plurality of color channels; andadapting weights of the AI-based denoiser based on the loss.2.The method of Claim 1, wherein:the noisy training image and the ground truth image have an identical image color filter array (CFA) pattern; andthe identical image CFA pattern comprises a Tetra CFA pattern, a Hexa-Deca CFA pattern, or a Nona CFA pattern.3.The method of any one of Claims 1 and 2, wherein determining the loss comprises using a linear combination of a mean absolute error (L1) loss, a multi-scale structural similarity loss, and an inter-channel loss.4.The method of any one of Claims 1 to 3, wherein generating the noisy ground truth image comprises adding zero-mean Gaussian noise to the ground truth image.5.The method of any one of Claims 1 to 4, wherein the noisy training image comprises an exposure value zero (EV0) image or a lower exposure value image.6.The method of any one of Claims 1 to 5, wherein determining the loss comprises using a total variation loss between the denoised images and a corresponding color channel for the ground truth image.7.The method of any one of Claims 1 to 6, wherein the noisy training image comprises one of a plurality of noisy training images, each noisy training image having a different proportion of noise.8.An electronic device comprising:at least one processing device configured to train an artificial intelligence (AI)-based denoiser;wherein, to train the AI-based denoiser, the at least one processing device is configured to:obtain a noisy training image and a ground truth image;convert the noisy training image into a first plurality of color channels;generate, using the AI-based denoiser, denoised images from the first plurality of color channels, wherein each denoised image corresponds to a respective color channel of the first plurality of color channels;generate a noisy ground truth image from the ground truth image;convert the noisy ground truth image into a second plurality of color channels;determine a loss based on a comparison between the denoised images and the second plurality of color channels; andadapt weights of the AI-based denoiser based on the loss.9.The electronic device of Claim 8, wherein:the noisy training image and the ground truth image have an identical image color filter array (CFA) pattern; andthe identical image CFA pattern comprises a Tetra CFA pattern, a Hexa-Deca CFA pattern, or a Nona CFA pattern.10.The electronic device of any one of Claims 8 and 9, wherein, to determine the loss, the at least one processing device is configured to use a linear combination of a mean absolute error (L1) loss, a multi-scale structural similarity loss, and an inter-channel loss.11.The electronic device of any one of Claims 8 to 10, wherein, to generate the noisy ground truth image, the at least one processing device is configured to add zero-mean Gaussian noise to the ground truth image.12.The electronic device of any one of Claims 8 to 11, wherein the noisy training image comprises an exposure value zero (EV0) image or a lower exposure value image.13.The electronic device of any one of Claims 8 to 12, wherein, to determine the loss, the at least one processing device is configured to use a total variation loss between the denoised images and a corresponding color channel for the ground truth image.14.The electronic device of any one of Claims 8 to 13, wherein the noisy training image comprises one of a plurality of noisy training images, each noisy training image having a different proportion of noise.15.A computer-readable storage medium storing instructions, wherein the instructions, when executed by at least one processor, cause the at least one processor to perform the method of any one of Claims 1 to 7.

Images