Abstract: In a collaborative video processing method and system, a high resolution video input is optionally downscaled to a low resolution video using a down-sampling filter, followed by an end-to-end video coding system to encode the low resolution video for streaming over the Internet. The original high resolution is obtained at the client end by upscaling the low resolution video using a deep learning based high resolution scaling model, which can be trained in a pre-defined progressive order with low resolution videos having different compression parameters and downscaling factors.

Abstract: An apparatus for PVD is provided. The apparatus includes a chamber, a pedestal disposed in the chamber to accommodate a wafer, and a ring. The ring includes a ring body having a first top surface and a second top surface, and a barrier structure disposed between the first top surface and the second top surface. The barrier structure can further include at least a first portion and a second portion separated from each other. The second vertical distance is equal to or greater than the first vertical distance.

Abstract: A tethering device includes a primary housing supporting a retraction mechanism; a secondary housing configured to be connected to a tethered object; a retractable tether extending from the retraction mechanism to the secondary housing to movably couple the secondary housing to the primary housing; a first magnet assembly supported by the primary housing; and a second magnet assembly supported by the secondary housing, wherein the second magnet assembly is configured to engage with the first magnet assembly to bias the secondary housing towards the primary housing.

Abstract: A semiconductor device and method of forming the same is disclosed. The semiconductor device includes a semiconductor substrate, a first fin and a second fin extending from the semiconductor substrate, a first lower semiconductor feature directly over the first fin, and a second lower semiconductor feature directly over the second fin. Each of the first and second lower semiconductor features includes a top surface bending downward towards the semiconductor substrate. The semiconductor also further includes an upper semiconductor feature directly over and in physical contact with the first and second lower semiconductor features. The semiconductor device further includes a dielectric layer on sidewalls of the first and second lower semiconductor features.

Abstract: The invention discloses a system configured for assessing performance data, the system comprising a memory comprising instruction data representing a set of instructions; and a processor configured to communicate with the memory and to execute the set of instructions. The set of instructions, when executed by the processor, cause the processor to: acquire a plurality of performance data records associated with a performance indicator; classify each performance data record according to the performance indicator; identify a plurality of variables in the performance data records that contribute to the classifications; determine, based on the plurality of variables, a plurality of observations relating to the performance indicator and at least one variable of the plurality of variables; and deliver the plurality of observations for presentation to a user. A method is also disclosed.

Abstract: An imaging lens assembly includes a plurality of lens elements, wherein at least one of the lens elements is a dual molded lens element. The dual molded lens element includes a light transmitting portion and a light absorbing portion. The light transmitting portion includes an effective optical section. The light absorbing portion is located on at least one surface of an object-side surface and an image-side surface of the dual molded lens element, wherein a plastic material and a color of the light absorbing portion are different from a plastic material and a color of the light transmitting portion, and the light absorbing portion includes an opening. The opening is non-circular and disposed correspondingly to the effective optical section.

Abstract: A method includes forming a semiconductor fin protruding higher than top surfaces of isolation regions. A top portion of the semiconductor fin is formed of a first semiconductor material. A semiconductor cap layer is formed on a top surface and sidewalls of the semiconductor fin. The semiconductor cap layer is formed of a second semiconductor material different from the first semiconductor material. The method further includes forming a gate stack on the semiconductor cap layer, forming a gate spacer on a sidewall of the gate stack, etching a portion of the semiconductor fin on a side of the gate stack to form a first recess extending into the semiconductor fin, recessing the semiconductor cap layer to form a second recess directly underlying a portion of the gate spacer, and performing an epitaxy to grow an epitaxy region extending into both the first recess and the second recess.

Abstract: An embodiment of the disclosure provides a heater package including a substrate, a first barrier layer, at least one heater, and a second barrier layer. The first barrier layer is disposed on a surface of the substrate and has a first treatment layer on a side away from the substrate. The heater is disposed on the substrate and includes a heating layer and at least one electrode. The at least one electrode and the heating layer contact with each other. The second barrier layer covers an upper surface and a sidewall of the heater and has a second treatment layer on an opposite side or the side away from the substrate. A ratio of a thickness of the first treatment layer to a thickness of the first barrier layer and a ratio of a thickness of the second treatment layer to a thickness of the second barrier layer range from 0.03 to 0.2.

Abstract: The disclosure provides a bicycle front derailleur including a fixing component, a chain guiding component, a linkage assembly, a battery assembly, two terminals, a circuit board assembly and a driving module. Two opposite ends of the linkage assembly are respectively and pivotally disposed on the fixing component and the chain guiding component. The battery assembly includes a battery holder and a battery. The battery holder is disposed on the fixing component, and the battery holder has a battery slot. One terminal is disposed in the battery slot, and the battery is removably inserted into the battery slot and is electrically connected to the terminal. The circuit board assembly is disposed on the fixing component and electrically connected to the other terminal. The terminals are electrically connected to each other. The driving module is disposed on the fixing component and electrically connected to the circuit board assembly.

Abstract: A method includes polishing a semiconductor substrate of a first die to reveal first through-vias that extend into the semiconductor substrate, forming a dielectric layer on the semiconductor substrate, and forming a plurality of bond pads in the dielectric layer. The plurality of bond pads include active bond pads and dummy bond pads. The active bond pads are electrically coupled to the first through-vias. The first die is bonded to a second die, and both of the active bond pads and the dummy bond pads are bonded to corresponding bond pads in the second die.

Abstract: A 3D printing device having a main body, a forming platform, an infrared thermometer, a heater, a thermocouple and a calibrator is provided. A forming chamber is defined in the main body. The forming platform is accommodated in the forming chamber, and a top surface thereof forms a bottom of the forming chamber. A reference surface is defined in the forming chamber, and the reference surface has an infrared radiance approximate to the powder. The infrared thermometer in the main body is disposed above and toward the reference surface. The heater is disposed in the main body to conductively heat the reference surface. The thermocouple is arranged adjacent to the reference surface. The calibrator is electrically connected to the infrared thermometer and the thermocouple to compare respective measured temperatures of the reference surface, and thereby determine if the infrared thermometer should be cleared.

Abstract: An etchant is utilized to remove a semiconductor material. In some embodiments an oxidizer is added to the etchant in order to react with surrounding semiconductor material and form a protective layer. The protective layer is utilized to help prevent damage that could occur from the other components within the etchant.

Abstract: A plastic lens assembly includes at least two plastic lens elements and at least one cementing glue coating. The plastic lens elements include a first plastic lens element and a second plastic lens element. The first plastic lens element has a first optical effective portion and a first peripheral portion, wherein the first peripheral portion surrounds the first optical effective portion. The second plastic lens element has a second optical effective portion and a second peripheral portion, wherein the second peripheral portion surrounds the second optical effective portion. The cementing glue coating is disposed between the first optical effective portion and the second optical effective portion, and for cementing the first plastic lens element and the second plastic lens element, wherein at least one optical gap is formed between the first optical effective portion and the second optical effective portion.

Abstract: An electronic package structure is provided, including a substrate with an electronic component, an antenna element and a shielding element disposed on the substrate. The shielding element is positioned between the antenna element and the electronic component to prevent electromagnetic interference (EMI) from occurring between the antenna element and the electronic component. A method for fabricating the electronic package structure is also provided.

Abstract: The present disclosure provides a polypeptide including an anti-fibrin antibody and a serine protease moiety of human tissue plasminogen activator. Methods for treating thrombosis in a subject in need of such treatment using such polypeptide are also disclosed.

Abstract: Structures and methods for trench isolation are disclosed. In one example, a silicon-on-insulator (SOI) structure is disclosed. The SOI structure includes: a substrate, a dielectric layer and a polysilicon region. The substrate includes: a handle layer, an insulation layer arranged over the handle layer, a buried layer arranged over the insulation layer, and a trench extending downward from an upper surface of the buried layer and terminating in the handle layer. The dielectric layer is located on a bottom surface of the trench and contacting the handle layer. The polysilicon region is located in the trench and contacting the dielectric layer.

Abstract: A headlight device is provided. The headlight device includes a main headlight module and a high-speed headlight module. The main headlight module is capable of generating a first lighting area having a light pattern that complies with the regulations of low beam headlamps. In addition, the main headlight module includes a main optical axis, and the main headlight module defines a horizontal plane and a vertical plane perpendicular to the horizontal plane, and the main optical axis is parallel to the horizontal plane and the vertical plane.

Abstract: In various examples, a generative model is used to synthesize datasets for use in training a downstream machine learning model to perform an associated task. The synthesized datasets may be generated by sampling a scene graph from a scene grammar—such as a probabilistic grammar—and applying the scene graph to the generative model to compute updated scene graphs more representative of object attribute distributions of real-world datasets. The downstream machine learning model may be validated against a real-world validation dataset, and the performance of the model on the real-world validation dataset may be used as an additional factor in further training or fine-tuning the generative model for generating the synthesized datasets specific to the task of the downstream machine learning model.

Abstract: Provided are systems and methods that perform multi-task and/or multi-sensor fusion for three-dimensional object detection in furtherance of, for example, autonomous vehicle perception and control. In particular, according to one aspect of the present disclosure, example systems and methods described herein exploit simultaneous training of a machine-learned model ensemble relative to multiple related tasks to learn to perform more accurate multi-sensor 3D object detection. For example, the present disclosure provides an end-to-end learnable architecture with multiple machine-learned models that interoperate to reason about 2D and/or 3D object detection as well as one or more auxiliary tasks. According to another aspect of the present disclosure, example systems and methods described herein can perform multi-sensor fusion (e.g.

Abstract: The present invention provides an antenna structure, an antenna device and a wireless localization method. The antenna structure includes a first radiation unit including a plurality of first connecting portions and a plurality of first annular radiation portions, a second radiation unit including a plurality of second connecting portions and a plurality of second annular radiation portions, a first conductive wire and a plurality of second conductive wires. A first end of each first annular radiation portion is connected to the first connecting portion, and the second end thereof extends towards the first end of the adjacent first annular radiation portion. A first end of each second annular radiation portion is connected to the second connecting portion, and the second end thereof extends towards the first end of the adjacent second annular radiation portion.