Abstract: A wafer processing method is provided. The method includes a device providing step, a first lifting step, a wafer placing step, a second lifting step, a soaking step, and a third lifting steps, and a spinning cleaning step. The device providing step includes providing a multifunctional single wafer immersion and spin cleaning device, the device has a spin drive device, a wafer turntable, and a wafer receiving tray. A soaking tank is formed on the wafer receiving tray, and a watertight contact gasket is disposed on the wafer receiving tray to contact the wafer water-tightly such that in the soaking step, an appropriate water level of the liquid medicine can be accumulated to fully soak the wafer.

Abstract: A method and an apparatus for video processing are provided. The method includes that a decoding terminal receives a plurality of coded video frames coded using one or more generative adversarial networks (GANs), receives network parameters related to the one or more GANs, and decodes the plurality of coded video frames using GANs based on the network parameters. Further, the one or more GANs respectively implement one or more video coding functions including reference-frame coding, motion-compensated frame prediction, and residue-frame coding.

Abstract: Various embodiments provide for quantizing a trained neural network with removal of normalization with respect to at least one layer of the quantized neural network, such as a quantized multiple fan-in layer (e.g., element-wise add or sum layer).

Abstract: Quantized softmax layers in neural networks are described. Some embodiments involve receiving, at an input to a softmax layer of a neural network from an intermediate layer of the neural network, a non-normalized output comprising a plurality of intermediate network decision values. Then for each intermediate network decision value of the plurality of intermediate network decision values, the embodiment involves: calculating a difference between the intermediate network decision value and a maximum network decision value; requesting, from a lookup table, a corresponding lookup table value using the difference between the intermediate network decision value and the maximum network decision value; and selecting the corresponding lookup table value as a corresponding decision value. A normalized output is then generated comprising the corresponding lookup table value for said each intermediate network decision value of the plurality of intermediate network decision values.

Abstract: Aspects of the present disclosure address systems and methods for fixed-point quantization using a dynamic quantization level adjustment scheme. Consistent with some embodiments, a method comprises accessing a neural network comprising floating-point representations of filter weights corresponding to one or more convolution layers. The method further includes determining a peak value of interest from the filter weights and determining a quantization level for the filter weights based on a number of bits in a quantization scheme. The method further includes dynamically adjusting the quantization level based on one or more constraints. The method further includes determining a quantization scale of the filter weights based on the peak value of interest and the adjusted quantization level. The method further includes quantizing the floating-point representations of the filter weights using the quantization scale to generate fixed-point representations of the filter weights.

Abstract: A method and an apparatus for video processing are provided. The method includes that a decoding terminal receives a plurality of coded video frames coded using one or more generative adversarial networks (GANs), receives network parameters related to the one or more GANs, and decodes the plurality of coded video frames using GANs based on the network parameters. Further, the one or more GANs respectively implement one or more video coding functions including reference-frame coding, motion-compensated frame prediction, and residue-frame coding.

Abstract: Methods and apparatuses are provided for temporal profiling for neural network quantization. The method includes: obtaining a neural network that comprises anode connected to different paths at different time periods; obtaining node outputs for the node at the different time periods; determining statistic properties of the node outputs at the different time periods; and determining activation ranges of the node outputs based on the statistic properties.

Abstract: Methods and devices are provided for implementing efficient general deconvolution Implementation on hardware accelerator.

Abstract: A method to implement a fixed-point batchnorm layer in a neural network for data processing is provided in the present disclosure. The method includes: receiving fixed-point input data over a channel of a standalone floating-point batchnorm layer, and converting the floating-point input data into fixed-point input data of the standalone floating-point batchnorm layer; obtaining fixed-point quantization parameters in each channel based on the input data and floating-point parameters ?i, ?i, ?i in each channel; converting the standalone floating-point batchnorm layer based on the fixed-point quantization parameters into a fixed-point batchnorm layer for processing the fixed-point input data to generate fixed-point output data; and mapping the fixed-point batchnorm layer to a fixed-point convolution layer and the computation of convolution is done by matrix multiplication that can be executed on a GEMM engine.

Abstract: A method to implement a fixed-point scale layer in a neural network for data processing is provided in the present disclosure. The method includes: receiving fixed-point input data over a channel of a standalone floating-point scale layer, and converting the floating-point input data into fixed-point input data of the standalone floating-point scale layer; obtaining fixed-point quantization parameters in each channel based on the input data and floating-point parameters ?i, ?i in each channel; converting the standalone floating-point scale layer based on the fixed-point quantization parameters into a fixed-point scale layer for processing the fixed-point input data to generate fixed-point output data; and mapping the fixed-point scale layer to a fixed-point convolution layer and the computation of convolution is done by matrix multiplication that can be executed on a GEMM engine.

Abstract: A class-specific neural network for video compressed sensing and methods for training and testing the class-specific neural network are provided. The class-specific neural network includes a Gaussian-mixture model (GMM) and a plurality of encoders, where the GMM classifies video frame blocks with a plurality of clusters and assigns the video frame blocks to the plurality of clusters. Further, the plurality of encoders receive the video frame blocks and generate a plurality of compressed-sensed frame block vectors, where the plurality of encoders correspond to the plurality of clusters.

Abstract: A method for detecting moving objects in video frames, an apparatus and a non-transitory computer-readable storage medium thereof are provided. The method includes that: an encoder in a 3-dimenional (3D) separable convolutional neural network with multi-input multi-output (3DS_MM) receives a first input including multiple video frames, where the encoder includes a plurality of encoder layers including 3D separable convolutional neural network (CNN) layers; the encoder generates a first encoder output; and a decoder in the 3DS_MM receives the first encoder output and generates a first output including multiple first binary masks related to the first input, where the decoder includes a plurality of decoder layers comprising 3D separable transposed CNN layers.

Abstract: A system for evaluating a classifier of an image signal processor (ISP) includes (i) a microprocessor and (ii) memory storing training images, the microprocessor being capable of sending each training image to the ISP. The system includes machine-readable instructions stored within the memory and executed by the microprocessor capable of: (i) selecting a subset of images based upon a divider position, (ii) controlling the ISP to classify each image as belonging or not belonging to an object class, (iii) determining a positive-match count, (iv) determining an error count based upon the positive-match count and total number of training images belonging to the object class, (v) repeating, for other divider positions, steps of selecting, controlling, and determining to identify an optimal divider position and a minimum-error count; and (vi) determining the classifier's optimality by comparing the optimal divider position to a predetermined optimal divider position and a predetermined minimum-error count.

Abstract: Embodiments of an apparatus and process are described. The process includes capturing a first image frame using an image sensor, the image sensor including a pixel array comprising a plurality of pixels arranged in rows and columns and a color filter array optically coupled to the pixel array. A region of interest within the first image frame is determined, and the exposure time of the image sensor is adjusted to eliminate a substantial fraction of the visible light captured by the image sensor. A rolling shutter procedure is used with the pixel array to capture at least one subsequent frame using the adjusted exposure time, and a source of invisible radiation is activated while the rolling shutter enters the region of interest and deactivated when the rolling shutter exits the region of interest. Finally, an image of the region of interest is output. Other embodiments are disclosed and claimed.

Abstract: A system for evaluating a classifier of an image signal processor (ISP) includes (i) a microprocessor and (ii) memory storing training images, the microprocessor being capable of sending each training image to the ISP. The system includes machine-readable instructions stored within the memory and executed by the microprocessor capable of: (i) selecting a subset of images based upon a divider position, (ii) controlling the ISP to classify each image as belonging or not belonging to an object class, (iii) determining a positive-match count, (iv) determining an error count based upon the positive-match count and total number of training images belonging to the object class, (v) repeating, for other divider positions, steps of selecting, controlling, and determining to identify an optimal divider position and a minimum-error count; and (vi) determining the classifier's optimality by comparing the optimal divider position to a predetermined optimal divider position and a predetermined minimum-error count.

Abstract: An imaging system with single-photon-avalanche-diodes (SPADs) and sensor translation for capturing a plurality of first images to enable generation of an enhanced-resolution image includes (a) an image sensor with SPAD pixels for capturing the plurality of first images at a plurality of spatially shifted positions of the image sensor, respectively, and (b) an actuator for translating the image sensor, parallel to its light receiving surface, to place the image sensor at the plurality of spatially shifted positions. A method for capturing a plurality of first images that enable composition of an enhanced-resolution image includes (a) translating an image sensor parallel to its light receiving surface to place the image sensor at a plurality of spatially shifted positions, and (b) capturing, using SPAD pixels implemented in pixel array of the image sensor, the plurality of first images at the plurality of spatially shifted positions, respectively.

Abstract: An imaging system with single-photon-avalanche-diodes (SPADs) and sensor translation for capturing a plurality of first images to enable generation of an enhanced-resolution image includes (a) an image sensor with SPAD pixels for capturing the plurality of first images at a plurality of spatially shifted positions of the image sensor, respectively, and (b) an actuator for translating the image sensor, parallel to its light receiving surface, to place the image sensor at the plurality of spatially shifted positions. A method for capturing a plurality of first images that enable composition of an enhanced-resolution image includes (a) translating an image sensor parallel to its light receiving surface to place the image sensor at a plurality of spatially shifted positions, and (b) capturing, using SPAD pixels implemented in pixel array of the image sensor, the plurality of first images at the plurality of spatially shifted positions, respectively.

Abstract: Embodiments of an apparatus and process are described. The process includes capturing a first image frame using an image sensor, the image sensor including a pixel array comprising a plurality of pixels arranged in rows and columns and a color filter array optically coupled to the pixel array. A region of interest within the first image frame is determined, and the exposure time of the image sensor is adjusted to eliminate a substantial fraction of the visible light captured by the image sensor. A rolling shutter procedure is used with the pixel array to capture at least one subsequent frame using the adjusted exposure time, and a source of invisible radiation is activated while the rolling shutter enters the region of interest and deactivated when the rolling shutter exits the region of interest. Finally, an image of the region of interest is output. Other embodiments are disclosed and claimed.

Abstract: A feature detection process includes identifying an approximate location of a feature in a preliminary image. A gradient phase map of image pixel intensities within the approximate location is computed. A projection result is determined by applying a projection function to the gradient phase map. The projection result is analyzed to determine a state of the feature.

Abstract: A feature detection process includes identifying an approximate location of a feature in a preliminary image. A gradient phase map of image pixel intensities within the approximate location is computed. A projection result is determined by applying a projection function to the gradient phase map. The projection result is analyzed to determine a state of the feature.