Abstract: A self-tuning model compression methodology for reconfiguring a Deep Neural Network (DNN) includes: receiving a pre-trained DNN model and a data set; performing an inter-layer sparsity analysis to generate a first sparsity result; and performing an intra-layer sparsity analysis to generate a second sparsity result, including: defining a plurality of sparsity metrics for the network; performing forward and backward passes to collect data corresponding to the sparsity metrics; using the collected data to calculate values for the defined sparsity metrics; and visualizing the calculated values using at least a histogram.

Abstract: A k-cluster residue number system includes a processor and a memory. The processor is used to generate a modular set composed of p coprime integers, generate a dynamic range by taking a product of the p coprime integers, generate row indices for all integers in the dynamic range, generate column indices for all integers in the dynamic range, and generate a look-up table according to the row indices, the column indices and all integers in the dynamic set. The memory is used to store the look-up table. The p coprime integers include 2.

Abstract: A 3D integrated circuit includes a substrate, a first layer on top of the substrate, and a second layer on top of the first layer. The first layer includes a first chip, and a first network bridge formed at a first side of the first chip. The second layer includes a second chip, and a second network bridge formed at a first side of the second chip. The first chip and the first network bridge are coupled to the substrate through bumps. The second chip is coupled to the first chip and the first network bridge through bumps. The second network bridge is coupled to the first network bridge through bumps. The first network bridge and the second network bridge each include a network switch for controlling data transfer and/or power distribution.

Abstract: A method and system of virtual footwear try-on with improved occlusion processing are disclosed. The method comprises: capturing a foot image of a user; defining an ankle joint point in the foot image and defining at least one reference point according to the ankle joint point; capturing depth point data within a spatial range around the reference point and constructing an occlusion model according to the depth point data; positioning the occlusion model to a position corresponding to the foot image to obtain an occlusion processed image; positioning the footwear model on the occlusion processed image to produce a matching image of the footwear model and the foot; and hiding the occlusion model.

Abstract: A pooling operation method for a convolutional neural network includes the following steps of: reading multiple new data in at least one current column of a pooling window; performing a first pooling operation with the new data to generate at least a current column pooling result; storing the current column pooling result in a buffer; and performing a second pooling operation with the current column pooling result and at least a preceding column pooling result stored in the buffer to generate a pooling result of the pooling window. The first pooling operation and the second pooling operation are forward max pooling operations.

Abstract: A convolution operation device includes a convolution calculation module, a memory and a buffer device. The convolution calculation module has a plurality of convolution units, and each convolution unit performs a convolution operation according to a filter and a plurality of current data, and leaves a part of the current data after the convolution operation. The buffer device is coupled to the memory and the convolution calculation module for retrieving a plurality of new data from the memory and inputting the new data to each of the convolution units. The new data are not a duplicate of the current data. A convolution operation method is also disclosed.

Abstract: A face identification system includes a transmitter, a receiver, a database, an artificial intelligence chip, and a main processor. The transmitter is used for emitting at least one first light signal to an object. The receiver is used for receiving at least one second light signal reflected by the object. The database is used for saving training data. The artificial intelligence chip is coupled to the transmitter, the receiver, and the database for identifying a face image from the object according to the at least one second light signal and the training data. The main processor is coupled to the artificial intelligence chip for receiving a face identification signal generated from the artificial intelligence chip.

Abstract: An image depth decoder includes an NIR image buffer, a reference image ring buffer and a pattern matching engine. The NIR image buffer stores an NIR image inputted by a stream. The reference image ring buffer stores a reference image inputted by a stream. The pattern matching engine is coupled to the NIR image buffer and the reference image ring buffer, and performs a depth computation according to the NIR image and the reference image to output at least one depth value.

Abstract: A low-power face identification method includes detecting an object image, extracting two-dimensional image information of the object image, and when the two-dimensional image information is undetected as a face feature, disabling all related components for inhibiting a three-dimensional face recognition function.

Abstract: A multi-layer artificial neural network having at least one high-speed communication interface and N computational layers is provided. N is an integer larger than 1. The N computational layers are serially connected via the at least one high-speed communication interface. Each of the N computational layers respectively includes a computation circuit and a local memory. The local memory is configured to store input data and learnable parameters for the computation circuit. The computation circuit in the ith computational layer provides its computation results, via the at least one high-speed communication interface, to the local memory in the (i+1)th computational layer as the input data for the computation circuit in the (i+1)th computational layer, wherein i is an integer index ranging from 1 to (N?1).

Abstract: A method for compressing multiple original convolution parameters into a convolution operation chip includes steps of: determining a range of the original convolution parameters; setting an effective bit number for the range; setting a representative value, wherein the representative value is within the range; calculating differential values between the original convolution parameters and the representative value; quantifying the differential values to a minimum effective bit to obtain a plurality of compressed convolution parameters; and transmitting the effective bit number, the representative value and the compressed convolution parameters to the convolution operation chip.

Abstract: A convolution operation method includes the following steps of: performing convolution operations for data inputted in channels, respectively, so as to output a plurality of convolution results; and alternately summing the convolution results of the channels in order so as to output a sum result. A convolution operation device executing the convolution operation method is also disclosed.

Abstract: A buffer device includes input lines, an input buffer unit and a remapping unit. The input lines are coupled to a memory and configured to be inputted with data from the memory in a current clock. The input buffer unit is coupled to the input lines and configured to buffer one part of the inputted data and output the part of the inputted data in a later clock. The remapping unit is coupled to the input lines and the input buffer unit, and configured to generate remap data for a convolution operation according to the data on the input lines and the output of the input buffer unit in the current clock. A convolution operation method for a data stream is also disclosed.