Abstract: A digital compute-in-memory (DCIM) system includes a first DCIM macro. The first DCIM macro includes a first memory cell array and a first arithmetic logic unit (ALU). The first memory cell array has N rows that are configured to store weight data of a neural network in a single weight data download session, wherein N is a positive integer not smaller than two. The first ALU is configured to receive a first activation input, and perform convolution operations upon the first activation input and a single row of weight data selected from the N rows of the first memory cell array to generate first convolution outputs.

Abstract: A digital compute-in-memory (DCIM) macro includes a memory cell array and an arithmetic logic unit (ALU). The memory cell array stores weight data of a neural network. The ALU receives parallel bits of a same input channel in an activation input, and generates a convolution computation output of the parallel bits and target weight data in the memory cell array.

Abstract: A method of compiling a deep learning model includes reading metadata from a compiled result, the metadata indicating a structure of the deep learning model corresponding to a low-level IR, receiving shape information of an input tensor of the deep learning model, determining a shape of an output tensor of a first computation operation of the computation operations based on the shape information of the input tensor of the deep learning model and the structure of the deep learning model, tiling the output tensor of the first computation operation into one or more tiles according to the shape of the output tensor of the first computation operation and hardware limitations of a processor executing the deep learning model, and patching one or more copies of a templated hardware command into executable hardware commands.

Abstract: A method of compiling a deep learning model includes reading metadata from a compiled result, the metadata indicating a structure of the deep learning model corresponding to a low-level IR, receiving shape information of an input tensor of the deep learning model, determining a shape of an output tensor of a first computation operation of the computation operations based on the shape information of the input tensor of the deep learning model and the structure of the deep learning model, tiling the output tensor of the first computation operation into one or more tiles according to the shape of the output tensor of the first computation operation and hardware limitations of a processor executing the deep learning model, and patching one or more copies of a templated hardware command into executable hardware commands.

Abstract: A system receives a neural network model that includes asymmetric operations. Each asymmetric operation includes one or more fixed-point operands that are asymmetrically-quantized from corresponding floating-point operands. The system compiles a given asymmetric operation of the neural network model into a symmetric operation that includes a combined bias value. A compiler computes the combined bias value is a constant by merging at least zero points of input and output of the given asymmetric operation. The system then generates a symmetric neural network model including the symmetric operation for inference hardware to execute in fixed-point arithmetic.

Abstract: Techniques pertaining to dynamic enablement, disablement and adjustment of offset of a virtual periodic timing control signal based on one or more predefined events are described. A method may determine whether a first predefined event is beginning. The method may also enable an offset of the virtual periodic timing control signal for synchronizing one or more first system modules in response to a determination that the first predefined event is beginning. The one or more first system modules may be configured to control one or more operations of one or more second system modules. The one or more second system modules may be configured to process one or more image frames. The method may further determine whether the first predefined event is ending. The method may additionally disable the offset in response to a determination that the first predefined event is ending.

Abstract: Concepts and examples pertaining to scenario-based policy for performance and power management in an electronic apparatus are described. A processor of an electronic apparatus determines whether an application currently executed on the electronic apparatus is of a predefined type of applications. The processor controls one or more aspects of operations of the electronic apparatus in executing the application responsive to the determining indicating the application being of the predefined type of applications.

Abstract: Methods and systems for improving images and video captured by a device is provided. The methods and systems may involve a virtual 3D model for a scene that is in view of a device's sensors. The virtual 3D model may be generated using the sensor information and be used to produce a 2D lighting image. The lighting image may be applied to a captured image of the scene to improve lighting of the captured image or video. A virtual light source may be implemented as part of the process, which if desired, can be moved, adjusted, or modified in the virtual 3D model to adjust the lighting image and consequently adjust alighted image or a final image.

Abstract: Techniques pertaining to dynamic enablement, disablement and adjustment of offset of a virtual periodic timing control signal based on one or more predefined events are described. A method may determine whether a first predefined event is beginning. The method may also enable an offset of the virtual periodic timing control signal for synchronizing one or more first system modules in response to a determination that the first predefined event is beginning. The one or more first system modules may be configured to control one or more operations of one or more second system modules. The one or more second system modules may be configured to process one or more image frames. The method may further determine whether the first predefined event is ending. The method may additionally disable the offset in response to a determination that the first predefined event is ending.

Abstract: A technique, as well as select implementations thereof, pertaining to smart frequency boost for graphics-processing hardware is described. A method may involve monitoring a queue of a plurality of graphics-related processes pending to be executed by a graphics-processing hardware to determine whether one or more predetermined conditions of the graphics-related processes in the queue are met. The one or more predetermined conditions may include an accumulation condition of the graphics-related processes in the queue. The method may also involve dynamically adjusting at least one operating parameter of the graphics-processing hardware in response to a determination that each of the one or more predetermined conditions of the graphics-related processes in the queue is met.

Abstract: A key mechanism (22) includes a key body (222), a resilient metal sheet (226), a circuit board (26). The circuit board (26) is placed between the key body (222) and the resilient metal sheet (226). The circuit board (26) includes a first surface (2622) facing the key body (222) and a second surface (2624) facing the resilient metal sheet (226). The first surface (2622) has a number of ground points connecting to the key body (222). The second surface (2624) has a switch unit corresponding to the resilient metal sheet (226).

Abstract: A portable electronic device includes a cover section (54), a body section (52), a magnet (60) and a hall sensor (70). The cover section has a printed circuit board secured therein. The body section has a connecting portion, and the body section is rotatably connected to the cover section with the connecting portion. The magnet is secured in the connecting portion, and the hall sensor is electrically attached to the printed circuit board. The magnet acts on the hall sensor according to relative rotation between the cover section and the body section, thereby switching the portable electronic device to a work mode when the cover section is opened away from the body section or a power save mode when the cover section is closed to the body section.

Abstract: A key mechanism (22) includes a key body (222), a resilient metal sheet (226), a circuit board (26). The circuit board (26) is placed between the key body (222) and the resilient metal sheet (226). The circuit board (26) includes a first surface (2622) facing the key body (222) and a second surface (2624) facing the resilient metal sheet (226). The first surface (2622) has a number of ground points connecting to the key body (222). The second surface (2624) has a switch unit corresponding to the resilient metal sheet (226).

Abstract: A portable electronic device includes a cover section (54), a body section (52), a magnet (60) and a hall sensor (70). The cover section has a printed circuit board secured therein. The body section has a connecting portion, and the body section is rotatably connected to the cover section with the connecting portion. The magnet is secured in the connecting portion, and the hall sensor is electrically attached to the printed circuit board. The magnet acts on the hall sensor according to relative rotation between the cover section and the body section, thereby switching the portable electronic device to a work mode when the cover section is opened away from the body section or a power save mode when the cover section is closed to the body section.

Abstract: A power saving method for a mobile phone includes the steps of: setting a standard brightness range, wherein the standard brightness range comprises a maximum value and a minimum value; opening a camera (11); measuring an ambient brightness; determining whether the measured ambient brightness is less than the minimum value; turning on a keypad LED (12) and decreasing the brightness of a liquid crystal module (13) if the measured ambient brightness is less than the minimum value; or turning off the keypad LED if the measured ambient brightness is not less than the minimum value; determining whether the measured ambient brightness is more than the maximum value; and increasing the brightness of the liquid crystal module if the measured ambient brightness is more than the maximum value. A power saving system for a mobile phone is also provided.