Abstract: A method and system is disclosed for tuning a machine learning classifier. An object class requirement may be provided and include rank thresholds. The object class requirements may also include a range goal that defines a minimum distance from the object the machine learning algorithm should not provide false positive results. A base classifier may be trained using a weighted loss function that includes one or more weight values that are computed using the one or more object class requirements. An output of the weighted loss function may be evaluated using an objective function which may be established using the one or more object class requirements. The one or more weights may also be re-tuned using the weighted loss function if the output of the weighted loss function does not converge within a predetermined loss threshold.

Abstract: The present disclosure provides a computer implemented method for generating a pollution event report. The method comprises: receiving pollutant data in respect of a pollutant measured at a site; determining, based on the pollutant data, that a pollution event has occurred at the site; receiving, based on the determination that a pollution event has occurred at the site, audio and/or image data captured at the site; and generating a pollution event report including an association between the pollution event and the audio and/or image data. A corresponding computer program, computer readable storage medium and system are also provided.

Abstract: Systems and methods for six-degree of freedom (6-DoF) pose estimation of a user input device, e.g., in a three-dimensional (3D) display system rendering interactive augmented reality (AR) and/or virtual reality (VR) experiences include the user input device capturing, via a camera disposed at a forward-facing tip of the user input device, images in a direction that the user input device is directed and determining, via an inertial measurement unit (IMU), motion of the user input device in three-dimensional (3D) space. The user input device may then determine pose information associated with the user input device based on the images and motion of the user input device. The determination of the pose information may be via usage of at least one of a neural network model, estimation model trained on a set of unique and identifiable patterns, and/or an estimation model trained on a dataset of images.

Abstract: Described herein is a mixed-flow fan assembly. The assembly comprises an impeller comprising a plurality of blades extending from a rotatable hub, and a shroud extending circumferentially around the impeller, wherein the shroud is secured to the plurality of blades, a casing with a bellmouth disposed circumferentially around the shroud, defining a cavity between the casing and the shroud, with flow control clearance gaps at a downstream end and an upstream end of the casing. The assembly further comprises one or more first circular ribs extending axially from the shroud within the cavity and substantially parallel to a rotational axis of the impeller, and a plurality of de-swirl vanes configured at predefined positions on an inner surface of the bellmouth.

Abstract: Systems and methods for six-degree of freedom (6-DoF) pose estimation of a user input device, e.g., in a three-dimensional (3D) system rendering interactive augmented reality (AR) and/or virtual reality (VR) experiences include the user input device capturing, via a camera disposed at a forward-facing tip of the user input device, images in a direction the user input device is directed and providing the images to a computer system. The user input device provides inertial measurement unit (IMU) data to the computer system as well. The computer system may then determine pose information associated with the user input device based on the images and IMU data of the user input device. The determination of the pose information may be via usage of at least one of a neural network model, estimation model trained on a set of unique and identifiable patterns, and/or an estimation model trained on a dataset of images.

Abstract: Systems and methods for six-degree of freedom (6-DoF) pose estimation of a user input device, e.g., in a three-dimensional (3D) system rendering interactive augmented reality (AR) and/or virtual reality (VR) experiences include the user input device capturing, via a camera disposed at a forward-facing tip of the user input device, images in a direction the user input device is directed and providing the images to a computer system. The user input device provides inertial measurement unit (IMU) data to the computer system as well. The computer system may then determine pose information associated with the user input device based on the images and IMU data of the user input device. The determination of the pose information may be via usage of at least one of a neural network model, estimation model trained on a set of unique and identifiable patterns, and/or an estimation model trained on a dataset of images.

Abstract: An air handling unit is disclosed, including a motor-driven fan with an annular outlet configured to move air through a housing of the AHU, a primary heat exchanger configured to cool or heat the air passing through the AHU, one or more heat elements placed downstream of the motor-driven fan, the one or more heat elements being configured to transfer heat to the airflow, and a heat shield placed between the motor-driven fan and the one or more heat elements, and configured to reduce heat transfer from the one or more heat elements upstream of the heat shield, and redistribute the airflow upstream and downstream of the heat shield, wherein the heat shield includes a distribution of openings configured to allow air to flow through the heat shield.

Abstract: Embodiments provide a method for obstacle avoidance in a mobile robot. Images of objects are captured with mobile robot image sensors with a floor level perspective. Data corresponding to the images is used to train a neural network. The identified objects are classified by indicating whether the objects are a potential hazard. In one embodiment, an image sensor on a robot provides intensity and depth data for each of a plurality of pixels of the images. The intensity and depth data are fused to produce fused data. The fused data is then used to train the neural network.

Abstract: A syringe assembly including: (1) a syringe barrel defining a chamber; (2) a plunger mounted in the chamber and moveable with respect to the barrel; and (3) a cap holder assembly containing a cap and an absorbent material removably attached to the syringe.

Abstract: A soot sensing system includes a soot sensor having a first element, and circuitry to an amount of soot accumulated on the element and to control heating of the element in response to the soot accumulation. An electrostatic repelling voltage (ERV) may be applied to a sensor/heater element(s) during a contamination prevention mode (CPM) to repel ash and reduce contamination of the sensor. A pulsed heating voltage (PHV) may be applied to the elements during the CPM and a pulsed ERV may be applied to the elements during the “off” period of the PHV. All voltage to the elements may be turned off during the CPM and the elements may be floating/ground. A PHV may be applied to the elements during the CPM and no voltage may be applied to the elements during the “off” period of the PHV. A heating voltage may be applied to the elements during a CPM corresponding to a cold start.

Abstract: Systems and methods for six-degree of freedom (6-DoF) pose estimation of a user input device, e.g., in a three-dimensional (3D) display system rendering interactive augmented reality (AR) and/or virtual reality (VR) experiences include the user input device capturing, via a camera disposed at a forward-facing tip of the user input device, images in a direction that the user input device is directed and determining, via an inertial measurement unit (IMU), motion of the user input device in three-dimensional (3D) space. The user input device may then determine pose information associated with the user input device based on the images and motion of the user input device. The determination of the pose information may be via usage of at least one of a neural network model, estimation model trained on a set of unique and identifiable patterns, and/or an estimation model trained on a dataset of images.

Abstract: Systems and methods for six-degree of freedom (6-DoF) pose estimation of a user input device, e.g., in a three-dimensional (3D) system rendering interactive augmented reality (AR) and/or virtual reality (VR) experiences include the user input device capturing, via a camera disposed at a forward-facing tip of the user input device, images in a direction the user input device is directed and providing the images to a computer system. The user input device provides inertial measurement unit (IMU) data to the computer system as well. The computer system may then determine pose information associated with the user input device based on the images and IMU data of the user input device. The determination of the pose information may be via usage of at least one of a neural network model, estimation model trained on a set of unique and identifiable patterns, and/or an estimation model trained on a dataset of images.

Abstract: An antiseptic medical cap that may include a base, a sidewall, and a chamber configured to receive a proximal end of a medical fluid port having an opening. The antiseptic medical cap, in some instances, may include an insert having a gas-releasing anti-microbial compound and/or include a nitric oxide-releasing system that is configured to be positioned at least partially through the opening of the medical fluid port when the antiseptic medical cap receives the proximal end of the medical fluid port into the chamber.

Abstract: A Person-to-person payments method and system may include receiving a request from a computer network to transfer funds from a first account to a second account, almost instantly. An alias (mobile number or email) is received and is associated with the second account and checked against a database. If the alias exists, multiple options may apply for transferring funds. Next, a secure party identifier is generated and the transfer may be completed. A Payments Switch Module may also communicate with a Third Party Payment Service Provider a request from a Sending or Receiving Financial Institution or party, to send or receive funds from a specific account held by the Third Party Payment Service Provider. This may occur on an adhoc, automatic, or scheduled basis enabling management of ‘personal cash concentration.’ This feature with the Third Party Payment Service Provider be characterized as a “Bring it Home” feature of the system.

Abstract: Systems and methods are disclosed for providing a subscription service.

Abstract: Disclosed are substituted quinazolinone compounds, pharmaceutical compositions comprising such compounds, and methods of using such compounds and compositions to treat diseases, disorders, or conditions such as those relating to unregulated protein function and/or levels.

Abstract: Disclosed herein are cryoablation devices and methods of using the same, including to ablate gallbladders. In certain implementations, the device is a catheter that has an expandable cryogen balloon for expansion within a gallbladder and at least one suction opening for applying suction within the gallbladder. According to other implementations, the device has a cryogen probe and a suction catheter slidably positioned on the probe.

Abstract: Embodiments provide a method for obstacle avoidance in a mobile robot. Images of objects are captured with mobile robot image sensors with a floor level perspective. Data corresponding to the images is used to train a neural network. The identified objects are classified by indicating whether the objects are a potential hazard. In one embodiment, an image sensor on a robot provides intensity and depth data for each of a plurality of pixels of the images. The intensity and depth data are fused to produce fused data. The fused data is then used to train the neural network.

Abstract: A Person-to-person payments method and system may include receiving a request from a computer network to transfer funds from a first account to a second account, almost instantly. An alias (mobile number or email) is received and is associated with the second account and checked against a database. If the alias exists, multiple options may apply for transferring funds. Next, a secure party identifier is generated and the transfer may be completed. A Payments Switch Module may also communicate with a Third Party Payment Service Provider a request from a Sending or Receiving Financial Institution or party, to send or receive funds from a specific account held by the Third Party Payment Service Provider. This may occur on an adhoc, automatic, or scheduled basis enabling management of ‘personal cash concentration.’ This feature with the Third Party Payment Service Provider be characterized as a “Bring it Home” feature of the system.

Abstract: Disclosed herein are cryoablation devices and methods of using the same, including to ablate gallbladders. In certain implementations, the device is a catheter that has an expandable cryogen balloon for expansion within a gallbladder and at least one suction opening for applying suction within the gallbladder. According to other implementations, the device has a cryogen probe and a suction catheter slidably positioned on the probe.