Abstract: To improve the accuracy and the speed of an autofocus method, using which a present best focal plane (13) may be found in an automated manner, which enables a best possible image quality for an object (3), which is located at a specific working distance (11) to an optical imaging system (1), it is provided that at least one parameter used during a z-scan (17) be adapted in an automated manner as a function of a presently set optical zoom level and/or a current estimated value of the working distance (11). During the z-scan (17), a present location of a focal plane (12) of the optical imaging system (1) is displaced within a scanning range (14) along an optical z-axis (8) of the imaging system (1), wherein the individual focal planes (12) are each evaluated to identify the best focal plane (13) among them.

Abstract: Certain aspects of the present disclosure provide techniques for adaptive sampling for rendering using deep learning. This includes receiving, at a sampler in a rendering pipeline, a plurality of rendered pixel data, wherein the sampler includes a first machine learning (ML) model. It further includes generating a sampling map for the rendering pipeline using the first ML model and the plurality of rendered pixel data, including predicting a plurality of pixel values in the sampling map based on a generated distribution of pixel values. It further includes rendering an image using the sampler, the sampling map, and a denoiser in the rendering pipeline.

Abstract: Techniques are disclosed for denoising videos. In some embodiments, video frames are denoised using a denoising model that includes an encoder-decoder architecture and attention modules. During training of the denoising model, the attention modules learn weightings to upweight certain dimensions of input features to help pixel registration, remove ghosting artifacts, and improve temporal consistency when the frames of a video are being denoised. The denoising model can also be used to train a student denoising model that has a same architecture as, but is smaller and faster than, the denoising model. After training, noisy video frames can be input into the denoising model and/or the student denoising model to generate corresponding denoised video frames.

Abstract: A computer-implemented method of changing a face within an output image or video frame that includes: receiving an input image that includes a face presenting a facial expression in a pose; processing the image with a neural network encoder to generate a latent space point that is an encoded representation of the image; decoding the latent space point to generate an initial output image in accordance with a desired facial identity but with the facial expression and pose of the face in the input image; identifying a feature of the facial expression in the initial output image to edit; applying an adjustment vector to a latent space point corresponding to the initial output image to generate an adjusted latent space point; and decoding the adjusted latent space point to generate an adjusted output image in accordance with the desired facial identity but with the facial expression and pose of the face in the input image altered in accordance with the adjustment vector

Abstract: A computer-implemented method of changing a face within an output image or video frame includes: receiving an input image that includes a face presenting a facial expression in a pose; separately encoding different portions of the image by, for each separately encoded portion, generating a latent space point of the portion, thereby generating a plurality of multi-dimensional vectors where each multi-dimensional vector is an encoded representation of a different portion of the input image; concatenating the plurality of multi-dimensional vectors into a combined latent space vector; and decoding the combined latent space vector to generate the output image in accordance with a desired facial identity but with the facial expression and pose of the face in the input image

Abstract: An endoscope arrangement (1) including a camera head (2) with at least one distal image sensor (4) and an image-processing unit (3) which is connected to the camera head (2) via a connecting line (14). The connecting line (14) has at least one optical line (15) for transmitting image information. The connecting line (14) additionally has an electrical line (16) for serially transmitting electrical signals, it being possible to transmit electrical control signals bidirectionally via this electrical line.

Abstract: A system includes a machine learning (ML) model-based video downsampler configured to receive an input video sequence having a first display resolution, and to map the input video sequence to a lower resolution video sequence having a second display resolution lower than the first display resolution. The system also includes a neural network-based (NN-based) proxy video codec configured to transform the lower resolution video sequence into a decoded proxy bitstream. In addition, the system includes an upsampler configured to produce an output video sequence using the decoded proxy bitstream.

Abstract: Techniques are disclosed for enhancing videos using a machine learning model that is a temporally-consistent transformer model. The machine learning model processes blocks of frames of a video in which the temporally first input video frame of each block of frames is a temporally second to last output video frame of a previous block of frames. After the machine learning model is trained, blocks of video frames, or features extracted from the video frames, can be warped using an optical flow technique and transformed using a wavelet transform technique. The transformed video frames are concatenated along a channel dimension and input into the machine learning model that generates corresponding processed video frames.

Abstract: A system includes a computing platform having a hardware processor and a memory storing a software code and a neural network (NN) having multiple layers including a last activation layer and a loss layer. The hardware processor executes the software code to identify different combinations of layers for testing the NN, each combination including candidate function(s) for the last activation layer and candidate function(s) for the loss layer. For each different combination, the software code configures the NN based on the combination, inputs, into the configured NN, a training dataset including multiple data objects, receives, from the configured NN, a classification of the data objects, and generates a performance assessment for the combination based on the classification. The software code determines a preferred combination of layers for the NN including selected candidate functions for the last activation layer and the loss layer, based on a comparison of the performance assessments.

Abstract: One embodiment of the present invention sets forth a technique for performing remastering of video content. The technique includes determining a first input frame corresponding to a first frame included in a first video and a first target frame corresponding to a second frame included in a second video based on one or more alignments between the first frame and the second frame. The technique also includes executing a machine learning model to convert the first input frame into a first output frame. The technique further includes training the machine learning model based on one or more losses associated with the first output frame and the first target frame.

Abstract: Particulate high-emissivity (high-?) refractory products include a mixture of (a) a particulate refractory base material which includes at least one particulate binder material, at least one particulate refractory raw material filler material and optionally at least one refractory additive; and (b) a high-? pigment in an amount sufficient to impart high-? properties to the refractory product when cured of at least 0.80. The high-? pigment is homogenously dispersed throughout the particulate refractory base material and is thereby less susceptible to loss of high-? properties over time. The particulate high-? products may be formed into an castable wet mix, an aqueous slurry or an insulating aqueous foam and cured so as to provide a component part of a high temperature refractory structure (e.g., the walls or ceiling of a refractory furnace) having high-? properties.

Abstract: A process for the microfluidic electrochemical synthesis of geminal dipseudohalide or halide-pseudohalide compounds comprising the steps of pumping a solution comprising a compound of Formula I into a microfluidic electrochemical reactor in the presence of a base, one of a halide or pseudohalide salt (MY), and a mediator; applying an electrical current through the microfluidic electrochemical reactor; and performing oxidative addition to create a geminal dipseudohalide or halide-pseudohalide compound of the general Formula II

Abstract: A sensor die may include a set of sensing elements and a test structure associated with determining a magnetic sensitivity of the set of sensing elements. The test structure includes a first test sensing element sensitive in a direction in a plane defined by a surface of the sensor die, a second test sensing element sensitive in the direction in the plane defined by the surface of the sensor die, and a wire on chip (WoC) associated with applying a magnetic field to the first test sensing element and the second test sensing element. The first test sensing element, the second test sensing element, and the WoC may be arranged such that, when current flows through the WoC, the first test sensing element senses a component of the magnetic field in the direction, and the second test sensing element senses a component of the magnetic field in a perpendicular direction.

Abstract: The invention relates to a mixer for an exhaust gas system for mixing an additive into an exhaust gas flow of an internal combustion engine, having a first shell and at least a second shell which are arranged successively in the circumferential direction in relation to a center axis, each shell having at least two shell edges that are arranged offset in the circumferential direction and which each form a flow edge, wherein the flow edges of two circumferentially adjacent shell edges of two different shells delimit an inflow opening, such that at least one pipe end arranged coaxially with the center axis is provided with a circumferential pipe profile that has a nominal radius Rn and is used for connection to an exhaust pipe, the pipe end being formed by the circumferentially adjacent shells.

Abstract: Provided herein are Platelet Activating Factor 4 (PF4) polypeptide and other compositions and methods for improving cognitive function in an individual comprising treatment with PF4 and other polypeptides.

Abstract: A system includes a computing platform having a hardware processor and a memory storing a software code and a neural network (NN) having multiple layers including a last activation layer and a loss layer. The hardware processor executes the software code to identify different combinations of layers for testing the NN, each combination including candidate function(s) for the last activation layer and candidate function(s) for the loss layer. For each different combination, the software code configures the NN based on the combination, inputs, into the configured NN, a training dataset including multiple data objects, receives, from the configured NN, a classification of the data objects, and generates a performance assessment for the combination based on the classification. The software code determines a preferred combination of layers for the NN including selected candidate functions for the last activation layer and the loss layer, based on a comparison of the performance assessments.

Abstract: Restoration methods and systems are disclosed for video remastering. Techniques disclosed include receiving a video sequence. For each frame of the video sequence, techniques disclosed include encoding, by a degradation encoder, a video content associated with the frame into a latent vector. The latent vector is a representation of the degradation present in the video content; the degradation present in the video content includes one or more degradation types. Based on the latent vector and the video content, techniques disclosed further include generating, by a backbone network, one or more feature maps, and, then, restoring the frame based on the one or more feature maps.

Abstract: A magnetic sensor configured to detect a rotation of an object includes at least one sensor element configured to generate at least one sensor signal based on a magnetic field that is modulated by the rotation of the object; a sensor circuit configured to generate a data transmission signal based on the at least one sensor signal, wherein the data transmission signal comprises a plurality of data transmission blocks; and a memory configured to store transmission block status information indicative of whether or not one of the plurality of data transmission blocks has been triggered for transmission. The sensor circuit is configured to detect an interrupt event that disrupts the transmission of the data transmission signal and avoid a complete loss of a data transmission block due to the detected interrupt event based on the transmission block status information present at a time the interrupt event is detected.

Abstract: To simplify the use of a functional scope of a medical image recording system, for example an endoscopy system, a method is proposed in which an image processing unit of the image recording system recognizes predefined image recording situations on the basis of an image sequence recorded using an image sensor of the image recording system and in response thereto proposes an adaptation to a user that results in an improved display and/or an improved recording of the image sequence in the respective recognized image recording situation. The adaptation can relate here to an algorithm, using which the image sequence is processed after the recording, and/or an image recording method currently used to generate the image sequence.

Abstract: The present invention relates to bulk metallic glass (BMG) structures used for hydrogen applications such as hydrogen infrastructure and vehicle applications. The BMG structure may include a main body with at least one opening, wherein the main body is made from a BMG material. The BMG structure may be configured to receive, store, transport and/or dispense hydrogen fuel in a form of fluid including one of gas, liquid, compressed gas or liquid, cryo-compressed hydrogen, and a combination thereof. The BMG structure may be configured to be under a direct exposure to hydrogen, wherein hydrogen is on the internal surface of the structure and is not on the exterior of the structure during operation.