Abstract: Disclosed are an image data processing method and a sensor device performing the same. The sensor device includes an image sensor configured to acquire image data, an image buffer configured to store the image data, and an image processor configured to generate image-processed data by applying a filter corresponding to a storage pattern of the image buffer to the image data stored in the image buffer.

Abstract: Disclosed is a method and apparatus for face detection using an adaptive threshold. The method includes determining a detection box in an input image, calculating a confidence score indicating whether an object in the detection box corresponds to a face, setting an adaptive threshold based on a size of the detection box, and determining whether the object in the detection box corresponds to a face based on comparing the confidence score to the adaptive threshold.

Abstract: A memory device and an operation method performed by the memory device are disclosed. The memory device includes a plurality of memories, and one or more memory banks including an in-memory operator configured to encode data stored in at least one of the plurality of memories, perform an assigned operation based on the encoded data, and decode the encoded data on which the operation is performed, and a memory controller configured to control the one or more memory banks.

Abstract: Disclosed is a method and apparatus for face detection using an adaptive threshold. The method includes determining a detection box in an input image, calculating a confidence score indicating whether an object in the detection box corresponds to a face, setting an adaptive threshold based on a size of the detection box, and determining whether the object in the detection box corresponds to a face based on comparing the confidence score to the adaptive threshold.

Abstract: Disclosed are a training method and apparatus for distributed training of a neural network, the training apparatus including processors configured to perform distributed training, wherein each of the processors is further configured to perform a forward direction operation for layers of the neural network, determine a loss of the neural network based on the forward direction operation, determine a local gradient for each layer of the neural network by performing a backward direction operation for the layers of the neural network based on the loss, determine whether to perform gradient clipping for a local gradient determined for a previous layer, in response to determining a local gradient for a current layer through the backward direction operation, determine an aggregated gradient based on the backward direction operation and the gradient clipping performed by each of the processors, and update parameters of the neural network based on the aggregated gradient.

Abstract: A processor-implemented method with data loading includes: dividing a training data set into a plurality of subsets based on sizes of a plurality of data files included in the training data set; loading, from each of the plurality of subsets, a portion of data files in the subset to a plurality of processors based on a proportion of a number of data files of the plurality of subsets in the subset and a batch size of distributed training; and reallocating, based on sizes of data files loaded to processors in a same group among the plurality of processors, the loaded data files to the processors in the same group.

Abstract: Disclosed are an image data processing method and a sensor device performing the same. The sensor device includes an image sensor configured to acquire image data, an image buffer configured to store the image data, and an image processor configured to generate image-processed data by applying a filter corresponding to a storage pattern of the image buffer to the image data stored in the image buffer.

Abstract: Disclosed are an image data processing method and a sensor device performing the same. The sensor device includes an image sensor configured to acquire image data, an image buffer configured to store the image data, and an image processor configured to generate image-processed data by applying a filter corresponding to a storage pattern of the image buffer to the image data stored in the image buffer.

Abstract: A processor-implemented neural network data processing method includes: determining a total number of either one of a first feature value and values less than or equal to the first feature value, in feature data output from a layer of a neural network; determining a quantization parameter based on the determined number; quantizing the feature data based on the determined quantization parameter; and inputting the quantized feature data to a another layer of the neural network connected to the layer.

Abstract: Disclosed are a neural network-based training method, inference method and apparatus. The neural network-based inference method includes receiving a quantization level for quantizing a weight of a neural network and an activation value that is processed by the neural network, receiving a weight quantized based on the quantization level, generating a quantized activation value by quantizing the activation value based on the quantization level, and performing inference based on the quantized weight and the quantized activation value.

Abstract: Disclosed are methods and apparatuses for processing a convolution operation on a layer in a neural network. The method includes extracting a first target feature vector from a target feature map, extracting a first weight vector matched with the first target feature vector from a first-type weight element, based on matching relationships for depth-wise convolution operations between target feature vectors of the target feature map and weight vectors of the first-type weight element, generating a first intermediate feature vector by performing multiplication between the first target feature vector and the first weight vector, generating a first hidden feature vector by accumulating the first intermediate feature vector and a second intermediate feature vector generated based on a second target feature vector, and generating a first output feature vector of an output feature map based on a point-wise convolution operation between the first hidden feature vector and a second-type weight element.

Abstract: Disclosed are an image data processing method and a sensor device performing the same. The sensor device includes an image sensor configured to acquire image data, an image buffer configured to store the image data, and an image processor configured to generate image-processed data by applying a filter corresponding to a storage pattern of the image buffer to the image data stored in the image buffer.

Abstract: A system includes: an image sensor configured to acquire an image; an image processor configured to generate a quantized image based on the acquired image using a trained quantization filter; and an output interface configured to output the quantized image.

Abstract: A processor-implemented neural network method includes: generating a first output line of an output feature map by performing a convolution operation between a first input line group of an input feature map and weight kernels; generating a first output of an operation block including the convolution operation based on the first output line; and storing the first output in a memory in which the input feature map is stored by overwriting the first output to a memory space of the memory.

Abstract: Disclosed is a method and apparatus for face detection using an adaptive threshold. The method includes determining a detection box in an input image, calculating a confidence score indicating whether an object in the detection box corresponds to a face, setting an adaptive threshold based on a size of the detection box, and determining whether the object in the detection box corresponds to a face based on comparing the confidence score to the adaptive threshold.

Abstract: A processor-implemented data processing method and apparatus for a neural network is provided. The data processing method includes generating cumulative data by accumulating results of multiplication operations between at least a portion of input elements in an input plane and at least a portion of weight elements in a weight plane, and generating an output plane corresponding to an output channel among output planes of an output feature map respectively corresponding to output channels based on the generated cumulative data.