Sensor data processing systems and techniques are described. In some examples, a sensor data processing system receives sensor data (e.g., image data). The system receives a sensor data processing parameter that is associated with a sensor data processing function, such as denoising, and that is consistent across the sensor data. The system adjusts a trained machine learning model based on the sensor data processing parameter to generate an adjusted machine learning model. The system processes the sensor data using the adjusted machine learning model to apply the sensor data processing function to the sensor data according to the sensor data processing parameter and to generate processed sensor data (e.g., denoised image data).

A computer-implemented method for color correction includes obtaining a noise evaluation image by adding noise to a noise-free image, color correcting the noise evaluation image using a plurality of color correction parameters, color correcting the noise-free image using the plurality of color correction parameters, determining a noise amplification metric at least by comparing the corrected noise evaluation image with the corrected noise-free image, and adjusting the plurality of color correction parameters based on the noise amplification metric.

A reproduction device is configured to receive light emitted based on a plurality of signals having a specific amplitude in a specific color space on which information is superimposed, and to reproduce the information based on the received light. The reproduction device includes: a memory; and a processor coupled to the memory and configured to: generate a plurality of signals in the specific color space from the received light, correct the generated plurality of signals based on the specific amplitude, and acquire the information based on the corrected plurality of signals.

Digital image processing circuitry converts images in a color filter array (CFA) color space to images in a luminance-chrominance (YUV) 4:2:0 color space, and the images in the YUV 4:2:0 color space are processed by the digital image processing circuitry in the YUV 4:2:0 color space, for example, to apply noise filtering, etc. The converting includes simultaneously receiving pixel data defining a macro-pixel in the CFA color space. The processing in the YUV color space is applied on a macro-pixel level to the macro-pixel of the image in the YUV color space.

Embodiments of the present disclosure relate to highlight recovery of a high-resolution image using a single low-resolution image captured at a lower exposure. An example apparatus includes a hue target circuit that receives an input image at a high-resolution including at least one pixel with a clipped color channel. For example, the input image is a Blue sky image with a pixel having clipped Blue channel. The hue target circuit also receives a set of candidate hue maps having a pixel resolution lower than the high-resolution of the input image. The hue target circuit generates a target hue value for the at least one pixel using the pixel information of the set of candidate hue maps. The apparatus also includes a hue recovery circuit that generates a recovered version of the input image by adjusting hue information of the clipped color channel based on the generated target hue.

An image processing apparatus for performing image processing for image data. The image processing apparatus includes a signal separation unit configured to separate a color signal and a luminance signal included in the image data, and a color signal control unit configured to control chroma of the color signal used for the image processing based on a luminance value of the luminance signal.

Various embodiments provide tone mapping of images and video from one dynamic range to an available dynamic range of a display device while preserving or enhancing image details. According to one embodiment, an enhanced tone mapping module is configured to decrease a luminance of an image or video from a high dynamic range to a standard dynamic range. Conversely, according to one embodiment, an enhanced inverse tone mapper to increase a luminance of an image or video from a standard dynamic range to a high dynamic range.

Obtaining color adjusted image portions using sub-three-dimensional look-up tables may include obtaining the color adjusted image portion by obtaining a value for an input image portion wherein the value includes a first parameter, a second parameter, and a third parameter, obtaining a color adjusted first parameter from at least a first one-dimensional look-up table based on the first parameter, the second parameter, and the third parameter, obtaining a color adjusted second parameter from at least a second one-dimensional look-up table based on the first parameter, the second parameter, and the third parameter, obtaining a color adjusted third parameter from at least a third one-dimensional look-up table based on the first parameter, the second parameter, and the third parameter, and outputting the color adjusted image portion based on a combination of the color adjusted first parameter, the color adjusted second parameter, and the color adjusted third parameter.

Various examples are directed to the generating and playback of panoramic videos comprising a plurality of tiles. Each tile may be a video having a tile field-of-view that is a portion of a full field-of-view of the panoramic video. An active field-of-view may be displayed from the first tile, where the active field-of-view is selected from a first tile field-of-view. A translate command may indicate a translate direction for the active field of view. A second tile may be requested. The second tile may have a second tile field-of-view that is adjacent to the first file field-of-view in the translate direction. The active field of view may be translated in the translate direction. When the active field-of-view reaches an edge of the first tile field-of-view, the active field-of-view may be displayed, at least in part, from the second tile.

A light locus of an imaging system is generated in a chromaticity space of two dimensions. The light locus represents a collection of candidate illuminants. The imaging system captures a gray-card image under each of N light sources to obtain N points in the chromaticity space, wherein N is a positive integer no less than three. Each point in the chromaticity space is described by a coordinate pair calculated from red (R), green (G) and blue (B) tristimulus values of the point. A second order polynomial function is calculated by curve-fitting the N points, and the light locus is generated to represent the second order polynomial in the chromaticity space. One of the candidate illuminants from the light locus is then identified as an illuminant for an image captured by the imaging system. A method for color transformation between two imaging systems in a chromaticity space is also described.