Abstract: The present disclosure relates to systems, methods, and computer-readable media for selectively identifying image frames from an input video to provide to an image processing model based on camera statistics. For example, systems disclosed herein include receiving an input video and associated camera statistics from a video capturing device. The systems disclosed herein further include identifying select image frames to provide to the image processing model based on the camera statistics and based on an application of the image processing model. The systems disclosed herein further include selectively identifying and providing camera statistics to the image processing model. By selectively providing data to the image processing model based on camera statistics, the systems disclosed herein can leverage capabilities of video capturing devices to significantly reduce the expense of processing resources when utilizing a variety of image processing models.

Abstract: The present disclosure relates to systems, methods, and computer-readable media for selectively identifying pixel data to provide as an input to an image processing model based on motion data associated with the content of a digital video. For example, systems disclosed herein include receiving a compressed digital video and decompressing the compressed digital video to generate a decompressed digital video. The systems disclosed herein further include extracting or otherwise identifying motion data while decompressing the compressed digital video. The systems disclosed herein also include analyzing the motion data to determine a subset of pixel data from the decompressed digital video to provide as input to an image processing model trained to generate an output based on input pixel data.

Abstract: The present disclosure relates to systems, methods, and computer-readable media for selectively identifying pixel data to provide as an input to an image processing model based on motion data associated with the content of a digital video. For example, systems disclosed herein include receiving a compressed digital video and decompressing the compressed digital video to generate a decompressed digital video. The systems disclosed herein further include extracting or otherwise identifying motion data while decompressing the compressed digital video. The systems disclosed herein also include analyzing the motion data to determine a subset of pixel data from the decompressed digital video to provide as input to an image processing model trained to generate an output based on input pixel data.

Abstract: The present disclosure relates to systems, methods, and computer-readable media for selectively identifying pixel data to provide as an input to an image processing model based on motion data associated with the content of a digital video. For example, systems disclosed herein include receiving a compressed digital video and decompressing the compressed digital video to generate a decompressed digital video. The systems disclosed herein further include extracting or otherwise identifying motion data while decompressing the compressed digital video. The systems disclosed herein also include analyzing the motion data to determine a subset of pixel data from the decompressed digital video to provide as input to an image processing model trained to generate an output based on input pixel data.

Abstract: The present disclosure relates to systems, methods, and computer-readable media for selectively identifying image frames from an input video to provide to an image processing model based on camera statistics. For example, systems disclosed herein include receiving an input video and associated camera statistics from a video capturing device. The systems disclosed herein further include identifying select image frames to provide to the image processing model based on the camera statistics and based on an application of the image processing model. The systems disclosed herein further include selectively identifying and providing camera statistics to the image processing model. By selectively providing data to the image processing model based on camera statistics, the systems disclosed herein can leverage capabilities of video capturing devices to significantly reduce the expense of processing resources when utilizing a variety of image processing models.

Abstract: Methods and devices for colorizing an infrared image stream may include receiving an infrared (IR) image stream of a scene. The methods and devices may include detecting at least one object in the IR image stream and receiving, from a red green blue (RGB) knowledge base, a RGB texture corresponding to the at least one object detected in the IR image stream. The methods and devices may include processing the RGB texture by matching coordinates in the RGB texture with corresponding coordinates in the at least one object to produce a colorized IR image stream. The methods and devices may include transmitting the colorized IR image stream of the scene wherein the at least one object is displayed in color based on the processing.

Abstract: Described are examples for generating a high dynamic range image from a plurality of images. A plurality of images of a real world scene can be obtained from one or more image sensors, wherein at least two of the plurality of images are captured based on different intensity parameters. Intensity information for the real world scene can be determined based at least in part on processing the at least two of the plurality of images. A high dynamic range image corresponding to the real world scene can be generated based at least in part on adding the intensity information to pixels of at least one of the plurality of images.

Abstract: Methods and devices for masking latency may include detecting a pause in receiving an image stream from an imaging device and generating one or more virtual image frames, each including a status indicator to indicate a status of the imaging device when the pause in receiving the image stream is detected. The methods and devices may also include generating, at the operating system, a data stream with the one or more virtual image frames inserted after a last image frame of the received image stream. In addition, the methods and devices may include transmitting the data stream to an application.

Abstract: Described are examples for generating a high dynamic range image from a plurality of images. A plurality of images of a real world scene can be obtained from one or more image sensors, wherein at least two of the plurality of images are captured based on different intensity parameters. Intensity information for the real world scene can be determined based at least in part on processing the at least two of the plurality of images. A high dynamic range image corresponding to the real world scene can be generated based at least in part on adding the intensity information to pixels of at least one of the plurality of images.

Abstract: Techniques for combining images to reduce motion blur in an output image are described. Data indicative of a first image and a second image of a scene is received. At least one portion of the first image associated with motion blur is identified. The portion of the first image is compared to a corresponding second portion of the second image. First pixels within the at least one portion of the first image and second pixels within the second portion of the second image not associated with motion blur are selected. The selected first pixels and second pixels are combined into an output image. The combined selected pixels are rendered on a display device.

Abstract: Described are examples for generating a high dynamic range image from a plurality of images. A plurality of images of a real world scene can be obtained from one or more image sensors, wherein at least two of the plurality of images are captured based on different intensity parameters. Intensity information for the real world scene can be determined based at least in part on processing the at least two of the plurality of images. A high dynamic range image corresponding to the real world scene can be generated based at least in part on adding the intensity information to pixels of at least one of the plurality of images.

Abstract: Described are examples for generating a high dynamic range image from a plurality of images. A plurality of images of a real world scene can be obtained from one or more image sensors, wherein at least two of the plurality of images are captured based on different intensity parameters. Intensity information for the real world scene can be determined based at least in part on processing the at least two of the plurality of images. A high dynamic range image corresponding to the real world scene can be generated based at least in part on adding the intensity information to pixels of at least one of the plurality of images.

Abstract: Methods and devices for masking latency may include detecting a pause in receiving an image stream from an imaging device and generating one or more virtual image frames, each including a status indicator to indicate a status of the imaging device when the pause in receiving the image stream is detected. The methods and devices may also include generating, at the operating system, a data stream with the one or more virtual image frames inserted after a last image frame of the received image stream. In addition, the methods and devices may include transmitting the data stream to an application.

Abstract: Methods and devices for colorizing an infrared image stream may include receiving an infrared (IR) image stream of a scene. The methods and devices may include detecting at least one object in the IR image stream and receiving, from a red green blue (RGB) knowledge base, a RGB texture corresponding to the at least one object detected in the IR image stream. The methods and devices may include processing the RGB texture by matching coordinates in the RGB texture with corresponding coordinates in the at least one object to produce a colorized IR image stream. The methods and devices may include transmitting the colorized IR image stream of the scene wherein the at least one object is displayed in color based on the processing.

Abstract: Techniques for combining images to reduce motion blur in an output image are described. Data indicative of a first image and a second image of a scene is received. At least one portion of the first image associated with motion blur is identified. The portion of the first image is compared to a corresponding second portion of the second image. First pixels within the at least one portion of the first image and second pixels within the second portion of the second image not associated with motion blur are selected. The selected first pixels and second pixels are combined into an output image. The combined selected pixels are rendered on a display device.