Abstract: Provided are computing systems, methods, and platforms that train query processing models, such as large language models, to perform query intent classification tasks by using retrieval augmentation and multi-stage distillation. Unlabeled training examples of queries may be obtained, and a set of the training examples may be augmented with additional feature annotations to generate augmented training examples. A first query processing model may annotate the retrieval augmented queries to generate inferred labels for the augmented training examples. A second query processing model may be trained on the inferred labels, distilling the query processing model that was trained with retrieval augmentation into a non-retrieval augmented query processing model. The second query processing model may annotate the entire set of unlabeled training examples. Another stage of distillation may train a third query processing model using the entire set of unlabeled training examples without retrieval augmentation.

Abstract: A method for fluorescent imaging of a subject, the method having the steps of: placing an illuminator in a surgical area of a subject, the illuminator having an emission band substantially similar to an fluorescence imaging agent; administering the fluorescence imaging agent to the subject; acquiring a plurality of images of the surgical area of the subject showing fluorescence intensity over a predetermined time period; and analyzing the plurality of images to determine changes in sensed fluorescence from the fluorescence imaging agent; wherein an exposure gain is applied to the images based on changes in detected intensity of the fluorescence from the illuminator.

Abstract: A vehicular vision system includes an electronic control unit (ECU) disposed at a vehicle and a camera having a CMOS imaging sensor operable to capture image data. Image data captured by the imaging sensor of the camera is conveyed from the camera to the ECU via a single core coaxial cable. The camera is in bidirectional communication with the ECU over the single core coaxial cable. The single core coaxial cable commonly carries (i) image data captured by the imaging sensor for processing at a data processor of the ECU and (ii) power from a DC power supply of the ECU to the camera. Image data captured by the imaging sensor is serialized at a data serializer of the camera and is conveyed to the ECU via the single core coaxial cable and is deserialized at the ECU by a data deserializer of the ECU.

Abstract: An endoscopic camera system having an optical assembly; a first color image sensor transmitting a first low dynamic range image; a second color image sensor transmitting a second low dynamic range image; a monochrome image sensor transmitting a monochrome image; an HDR processor coupled to the first color image sensor and the second color image sensor for receiving the first low dynamic range image and the second low dynamic range image, the HDR processor being configured to combine the first low dynamic range image and the second dynamic range image into a high dynamic range image; and a combiner coupled to the HDR processor and the monochrome image sensor, the combiner being configured to combine the high dynamic range image with the monochrome image.

Abstract: A device and method for color correction of two camera systems each having an imaging device and an illumination source. The camera systems are separately white balanced with the imaging device outside a scene. Their white balance gains and color correction matrices are saved. Based on the measurements, combined white balance gains and combined color correction matrices are computed and saved. Thereafter, with the imaging devices in a scene, performing white balancing by measuring the R, G, and B values of both imaging devices together, and then white balancing by measuring the R, G, and B values of each separate imaging device with the other camera system's light off. Comparisons are made of each camera's scene measurements against the combined scene measurements. If they are significantly different, the combined light set of white balance gain and color correction matrix is applied to digital signal processing paths of each camera system.

Abstract: A vehicular vision system includes an electronic control unit (ECU) disposed at a vehicle and a camera having a CMOS imaging sensor operable to capture image data. Image data captured by the imaging sensor of the camera is conveyed from the camera to the ECU via a single core coaxial cable. The camera is in bidirectional communication with the ECU over the single core coaxial cable. The single core coaxial cable commonly carries (i) image data captured by the imaging sensor for processing at a data processor of the ECU and (ii) power from a DC power supply of the ECU to the camera. Image data captured by the imaging sensor is serialized at a data serializer of the camera and is conveyed to the ECU via the single core coaxial cable and is deserialized at the ECU by a data deserializer of the ECU.

Abstract: A device and method for color correction of two camera systems each having an imaging device and an illumination source. The camera systems are separately white balanced with the imaging device outside a scene. Their white balance gains and color correction matrices are saved. Based on the measurements, combined white balance gains and combined color correction matrices are computed and saved. Thereafter, with the imaging devices in a scene, performing white balancing by measuring the R, G, and B values of both imaging devices together, and then white balancing by measuring the R, G, and B values of each separate imaging device with the other camera system's light off. Comparisons are made of each camera's scene measurements against the combined scene measurements. If they are significantly different, the combined light set of white balance gain and color correction matrix is applied to digital signal processing paths of each camera system.

Abstract: A vehicular vision system includes an ECU disposed at a vehicle and a front camera having a CMOS imaging sensor operable to capture image data. Image data captured by the imaging sensor of the front camera is conveyed from the front camera to the ECU via a single core coaxial cable. The front camera is in bidirectional communication with the ECU over the single core coaxial cable. The single core coaxial cable commonly carries (i) image data captured by the imaging sensor for processing at a data processor of the ECU and (ii) power from a DC power supply of the ECU to the front camera. Image data captured by the imaging sensor is serialized at a data serializer of the front camera and is transmitted to the ECU via the single core coaxial cable and is deserialized at the ECU by a data deserializer of the ECU.

Abstract: An endoscopic camera system having an optical assembly; a first color image sensor transmitting a first low dynamic range image; a second color image sensor transmitting a second low dynamic range image; a monochrome image sensor transmitting a monochrome image; an HDR processor coupled to the first color image sensor and the second color image sensor for receiving the first low dynamic range image and the second low dynamic range image, the HDR processor being configured to combine the first low dynamic range image and the second dynamic range image into a high dynamic range image; and a combiner coupled to the HDR processor and the monochrome image sensor, the combiner being configured to combine the high dynamic range image with the monochrome image.

Abstract: An endoscopic camera system having a camera that captures and outputs an image; a camera controller coupled to the camera; a user input device coupled to the camera or the camera controller, wherein the user input device is usable to select a region of interest in the image, the region of interest being a sub-part of the image; and wherein the camera controller: computes a measured luminance value for the region of interest; and adjusts an exposure in response to a comparison of the measured luminance value with a target luminance value

Abstract: An endoscopic camera system having a camera that captures and outputs an image; a camera controller coupled to the camera; a user input device coupled to the camera or the camera controller, wherein the user input device is usable to select a region of interest in the image, the region of interest being a sub-part of the image; and wherein the camera controller: computes a measured luminance value for the region of interest; and adjusts an exposure in response to a comparison of the measured luminance value with a target luminance value.

Abstract: A vehicular vision system includes an ECU disposed at a vehicle and a front camera having a CMOS imaging sensor operable to capture image data. Image data captured by the imaging sensor of the front camera is conveyed from the front camera to the ECU via a single core coaxial cable. The front camera is in bidirectional communication with the ECU over the single core coaxial cable. The single core coaxial cable commonly carries (i) image data captured by the imaging sensor for processing at a data processor of the ECU and (ii) power from a DC power supply of the ECU to the front camera. Image data captured by the imaging sensor is serialized at a data serializer of the front camera and is transmitted to the ECU via the single core coaxial cable and is deserialized at the ECU by a data deserializer of the ECU.

Abstract: A vehicular multi-camera surround view system includes a plurality of cameras disposed at a vehicle and having respective exterior fields of view, each of the cameras capturing respective image data. The cameras are connected to an ECU via respective ones of a plurality of single core coaxial cables. The ECU is disposed at the vehicle and includes (i) a data processor and (ii) a DC power supply. Each of the cameras is in full duplex bidirectional communication with the ECU over the respective single core coaxial cable. Each single core coaxial cable commonly carries (i) image data from the respective camera to the ECU for processing and (ii) power from the DC power supply of the ECU to the respective camera. The ECU combines image data conveyed by the cameras to form composite video images and outputs the composite video images to a display device having a video display screen.

Abstract: A device and method for color correction of two camera systems each having an imaging device and an illumination source. The camera systems are separately white balanced with the imaging device outside a scene. Their white balance gains and color correction matrices are saved. Based on the measurements, combined white balance gains and combined color correction matrices are computed and saved. Thereafter, with the imaging devices in a scene, performing white balancing by measuring the R, G, and B values of both imaging devices together, and then white balancing by measuring the R, G, and B values of each separate imaging device with the other camera system's light off. Comparisons are made of each camera's scene measurements against the combined scene measurements. If they are significantly different, the combined light set of white balance gain and color correction matrix is applied to digital signal processing paths of each camera system.

Abstract: A device and method for color correction of two camera systems each having an imaging device and an illumination source. The camera systems are separately white balanced with the imaging device outside a scene. Their white balance gains and color correction matrices are saved. Based on the measurements, combined white balance gains and combined color correction matrices are computed and saved. Thereafter, with the imaging devices in a scene, performing white balancing by measuring the R, G, and B values of both imaging devices together, and then white balancing by measuring the R, G, and B values of each separate imaging device with the other camera system's light off. Comparisons are made of each camera's scene measurements against the combined scene measurements. If they are significantly different, the combined light set of white balance gain and color correction matrix is applied to digital signal processing paths of each camera system.

Abstract: A vehicular multi-camera surround view system includes a plurality of cameras disposed at a vehicle and having respective exterior fields of view, each of the cameras capturing respective image data. The cameras are connected to an ECU via respective ones of a plurality of single core coaxial cables. The ECU is disposed at the vehicle and includes (i) a data processor and (ii) a DC power supply. Each of the cameras is in full duplex bidirectional communication with the ECU over the respective single core coaxial cable. Each single core coaxial cable commonly carries (i) image data from the respective camera to the ECU for processing and (ii) power from the DC power supply of the ECU to the respective camera. The ECU combines image data conveyed by the cameras to form composite video images and outputs the composite video images to a display device having a video display screen.

Abstract: A vehicular vision system includes a plurality of imaging sensors disposed at a vehicle and having respective exterior fields of view, each of the imaging sensors capturing respective image data. The imaging sensors are connected to a control via respective ones of a plurality of single core coaxial cables. The control is disposed at the vehicle and includes (i) a data processor and (ii) a DC power supply. Each single core coaxial cable commonly carries (i) image data from the respective imaging sensor to the control for processing and (ii) power from the DC power supply of the control to the respective imaging sensor. Each of the imaging sensors is in bidirectional communication with the control over the respective single core coaxial cable, with signal attenuation being below 100 dB at a frequency of 1 GHz per 100 m length of the respective single core coaxial cable.

Abstract: A vehicular vision system includes a plurality of imaging sensors disposed at a vehicle and having respective exterior fields of view, each of the imaging sensors capturing respective image data. The imaging sensors are connected to a control via respective ones of a plurality of single core coaxial cables. The control is disposed at the vehicle and includes (i) a data processor and (ii) a DC power supply. Each single core coaxial cable commonly carries (i) image data from the respective imaging sensor to the control for processing and (ii) power from the DC power supply of the control to the respective imaging sensor. Each of the imaging sensors is in bidirectional communication with the control over the respective single core coaxial cable, with signal attenuation being below 100 dB at a frequency of 1 GHz per 100 m length of the respective single core coaxial cable.

Abstract: A vehicular vision system includes a plurality of imaging sensors disposed at a vehicle and having respective exterior fields of view, each of the imaging sensors capturing respective image data. The imaging sensors are connected to a control via respective ones of a plurality of single core coaxial cables. Each single core coaxial cable commonly carries (i) image data from the respective imaging sensor to the control for processing and (ii) power to the respective imaging sensor. Each imaging sensor is capable of communicating via a particular communication protocol. While each of the imaging sensors is in a respective initial mode and after the communication protocol is transmitted by the control to each of the imaging sensors, each of the imaging sensors communicates with the control in accordance with the communication protocol. Each of the single core coaxial cables provides bidirectional communication between the control and the respective imaging sensor.

Abstract: A vehicular vision system includes a plurality of imaging sensors disposed at a vehicle and having respective exterior fields of view, each of the imaging sensors capturing respective image data. The imaging sensors are connected to a control via respective ones of a plurality of single core coaxial cables. Each single core coaxial cable commonly carries (i) image data from the respective imaging sensor to the control for processing and (ii) power to the respective imaging sensor. Each imaging sensor is capable of communicating via a particular communication protocol. While each of the imaging sensors is in a respective initial mode and after the communication protocol is transmitted by the control to each of the imaging sensors, each of the imaging sensors communicates with the control in accordance with the communication protocol. Each of the single core coaxial cables provides bidirectional communication between the control and the respective imaging sensor.