Abstract: A yeast-containing dough that is used to form a leavened dough product having a desired degree of leavening is initially formed as two separate dough subcomponents that, when combined, leavens through a fermentation reaction process to form a dough having the desired degree of leavening. The first dough subcomponent includes a leavening activator, flour, salt and water, while the second dough component includes a biological leavening agent, flour, and water. When combined, the leavening agent reacts in the presence of the biological leavening activator through a fermentation process to form the dough having a desired degree of leavening, which can then be heated to a temperature exceeding a viability temperature of the leavening activator to terminate the fermentation process and form the leavened dough product. The dough can also be combined with at least one non-dough component to form the leavened dough product.

Abstract: An electronic apparatus, a camera optical-zoom method, a camera optical-zoom unit, and a memory are disclosed. The electronic apparatus may include a camera, a sensor, a processor, and a memory, where the memory stores a program instruction which, when executed by the processor, causes the processor to perform the following steps: receiving image data of a photographed object acquired by the camera; and integrating image data of a full pixel area corresponding to a zoom value of the camera according to a preset algorithm, and outputting the integrated image data according to a coordinate sequence of pixels on the sensor, where the full pixel area includes pixel areas of the sensor corresponding to the zoom value and pixel areas of the sensor corresponding to a value greater than the zoom value.

Abstract: Systems and methods related to serial communication devices are provided. An example integrated circuit (IC) device includes interface circuitry coupled to a two-wire serial communication bus having a serial clock (SCL) line and a serial data (SDA) bus line. The IC device further includes bus stuck recovery circuitry to monitor for a local SDA fault condition at the IC device based on a number of clock cycles during which an internal SDA signal (e.g., generated by the IC device) drives the SDA bus line to a first signal state, the clock cycles based on a clock signal received from the SCL line; and responsive to the local SDA fault condition, release the SDA bus line independent of the internal SDA signal, where the SDA bus line is in a second signal state different from the first signal state based on the release.

Abstract: Examples of a computer system for malware detection and recovery include a computer having a processor and a memory having untrusted partition having untrusted data files and a trusted partition connected to the untrusted partition and having a backup application. The trusted partition may not be modifiable by the untrusted partition. The computer may be configured to detect whether a malware attack of the untrusted partition has occurred. In accordance with detecting that the malware attack has not occurred, the computer creates a trusted backup by storing at least some untrusted data files to a backup memory that is not modifiable by the untrusted partition. In accordance with detecting that the malware attack has occurred, the computer may prevent the creation of an additional trusted backup and may restore compromised files from the trusted backup.

Abstract: Disclosed herein is a methodology for refining novel-class features in a few-shot learning scenario by fine-tuning the feature extractor by reducing both class-agnostic biases and class-specific biases. A distribution calibration module is used to reduce the class-agnostic bias by normalizing the overall feature distribution for novel classes and further reshaping the feature manifold for fast adaptation during fine-tuning. Selected sampling is used to reduce class-specific bias by augmenting more data for better estimation.

Abstract: Disclosed herein is a system and method of identifying new products on a retail shelf using a feature extractor trained to extract features from images of products on the shelf and output identifying information regarding the product in the product image. The extracted features are compared to extracted features in a product library and a best-fit is obtained. A new product is identified if the distance between the features of the product on the shelf and the features of the best-fit product from the product library are above a predetermined threshold.

Abstract: The present disclosure provides a spliced display screen, which includes a plurality of display modules arranged closely and one or more positioning mechanisms installed on each display module, each positioning mechanism includes: a positioning pin and a groove, the one or more positioning pins are clamped in the one or more grooves, the one or more positioning pins and the one or more grooves cooperate to restrict relative movement in the horizontal direction of two adjacent display modules; an elastic structure arranged corresponding to the positioning pin, one end being connected to the display module and the other end being connected to the positioning pin; in response to being subjected to an external force, the one or more positioning pins rotate along a predetermined direction relative to the one or more grooves and leave the one or more grooves; in response to being disengaged from the external force, the one or more positioning pins return to an initial position along a direction opposite to the pred

Abstract: Disclosed herein is a methodology implementing feature map-level data augmentation in a feature map. Two or more units in the feature map are selected and the values of locations in the two or more units are swapped among the two or more units. Value perturbations applied around local units in the feature map implicitly lead to an unused data augmentation at the image level.

Abstract: Disclosed herein is an improved few-shot detector which utilizes a dynamic semantic network which takes as input a language feature and generates trainable parameters for a visual network. The visual network takes a visual feature as input and generates a classification and localization of an object.

Abstract: Provided herein in the biomedical field are a pharmaceutical composition, use thereof for treating or suppressing a cancer, reducing its severity, lowering its risk or inhibiting its metastasis in an individual, and a method for treating or suppressing a cancer, reducing its severity, lowering its risk or inhibiting its metastasis in an individual. The method comprises administering to the individual a therapeutically effective amount of an ALK inhibitor, and a therapeutically effective amount of one or more other anticancer reagents. It also relates to a kit comprising an ALK inhibitor, and one or more other anticancer reagents.

Abstract: A backlight module and a display device. The backlight module includes: a back plate, wherein the back plate has a bottom plate and a plurality of side plates, the side plates cooperate with the bottom plate to form an accommodating space, and the side of each of the side plates facing the accommodating space is provided with a groove having an opening facing the accommodating space; and a mold frame, includes bezels in one-to-one correspondence with the plurality of side plates; in each group of a side plate and a bezel, the bezel includes a clamping part, a supporting part and a limiting part, the clamping part is located at a side of the limiting part away from the accommodating space, and is embedded into the groove to clamp with the groove.

Abstract: An electronic apparatus, a camera optical-zoom method, a camera optical-zoom unit, and a memory are disclosed. The electronic apparatus may include a camera, a sensor, a processor, and a memory, where the memory stores a program instruction which, when executed by the processor, causes the processor to perform the following steps: receiving image data of a photographed object acquired by the camera; and integrating image data of a full pixel area corresponding to a zoom value of the camera according to a preset algorithm, and outputting the integrated image data according to a coordinate sequence of pixels on the sensor, where the full pixel area includes pixel areas of the sensor corresponding to the zoom value and pixel areas of the sensor corresponding to a value greater than the zoom value.

Abstract: The present invention makes public a method and an apparatus for the revalorization of various organic waste, by means of physics and mechanics. The bubbles are produced and injected into the mixture in the sealed tank under depressure, then establish the movement of toric rotation in the vertical direction. The high-speed movement bubble carrying focused energy vigorously collide into the mixture, thus showing the Rheo-fluidizing effect (Shear thinning), decreasing the viscosity and depolymerizing long-chain molecules. This achieves active bacteria killing, separation, concentration and de-emulsification; the solution's advantage is applied to various organic residue, with short and efficient processing cycles, without producing unpleasant odors and is non-polluting, chemical free, at low temperature, low energy consumption, no GHG emission, preserving the microorganisms and nutrients of organic matter, and separating the pollutants such as the oils and fats.

Abstract: A device configured to transcribe an appearance of a human being, including a common housing holding an image capturing sensor, a computing device having a data processor, and a computer program product including a first machine learning model trained for detecting and labeling human beings, a second machine learning model trained for detecting appearances of human beings and a transcription module to transcribe the detected appearances of human beings to text.

Abstract: There is provided a device configured to transcribe an appearance of a human being, said device comprising a common housing holding an image capturing sensor, a computing device comprising a data processor, and a computer program product comprising a first machine learning model trained for detecting and labeling human beings, a second machine learning model trained for detecting appearances of human beings and a transcription module to transcribe the detected appearances of human beings to text.

Abstract: A display method, an electronic device, and a storage medium are provided. The method includes: detecting state information of a first object; if the state information indicates that the first object appears or disappears, determining whether a trigger condition is satisfied according to the state information; and if the trigger condition is satisfied, transforming display of a second object according to a trigger event corresponding to the trigger condition, where the second object is independent of the first object.

Abstract: Method and apparatus for object detection by a robot are provided. The method comprises analyzing using a set of trained detection models, one or more first images of an environment of the robot to detect one or more objects in the environment of the robot, generating at least one fine-tuned model by training one or more of the trained detection models in the set, wherein the training is based on a second image of the environment of the robot and annotations associated with the second image, wherein the annotations identify one or more objects in the second image, updating the set of trained detection models to include the generated at least one fine-tuned model, and analyzing using the updated set of trained detection models, one or more third images of the environment of the robot to detect one or more objects in the environment.

Abstract: Systems, methods, and computer program products are described that include obtaining, at a processor, a first image from an image capture device onboard a computing device, detecting, using the processor and at least one sensor, a device orientation of the computing device associated with capture of the first image, determining, based on the device orientation and a tracking stack associated with the computing device, a rotation angle in which to rotate the first image, rotating the first image to the rotation angle to generate a second image, and generating neural network based estimates associated with the first image and the second image.

Abstract: A backlight assembly and a display device are provided in the present disclosure. The backlight assembly includes an optical lamination layer and a die-casting back plate. The die-casting back plate defines an accommodation space for accommodating the optical lamination layer, the die-casting back plate includes an inner surface and an outer surface arranged opposite to each other, the inner surface is a surface facing a bottom surface of the optical lamination layer, a recess region is arranged on the inner surface, and a plurality of rivets configured to assemble the die-casting back plate onto an object to be assembled is arranged on the outer surface. The plurality of rivets is located in a region corresponding to the recess region and spaced apart from a boundary of the recess region by a spacing.

Abstract: Disclosed is a backlight module. The backlight module includes a back plate, a light guide plate, and a clamp; the back plate includes a back plate body, and a plurality of side plates disposed along a periphery of the back plate body and connected to the back plate body, and a plane of each side plate is intersected with a plane of the back plate body. The light guide plate is disposed on a side, connected to the side plate, of the back plate body, and a gap is present between the light guide plate and each of the plurality of side plates. The clamp is disposed in the gap between the light guide plate and at least part of the plurality of side plates, and interference-fitted with the gap. The clamp is resilient, and the clamp and the side plate are positioned by a projection-recess fitting structure.