724 results about "Encoder decoder" patented technology

Techniques and tools for encoding and decoding motion vector information for video images are described. For example, a video encoder yields an extended motion vector code by jointly coding, for a set of pixels, a switch code, motion vector information, and a terminal symbol indicating whether subsequent data is encoded for the set of pixels. In another aspect, an encoder/decoder selects motion vector predictors for macroblocks. In another aspect, a video encoder/decoder uses hybrid motion vector prediction. In another aspect, a video encoder/decoder signals a motion vector mode for a predicted image. In another aspect, a video decoder decodes a set of pixels by receiving an extended motion vector code, which reflects joint encoding of motion information together with intra/inter-coding information and a terminal symbol. The decoder determines whether subsequent data exists for the set of pixels based on e.g., the terminal symbol.

Techniques and tools for encoding and decoding predicted images in interlaced video are described. For example, a video encoder or decoder computes a motion vector predictor for a motion vector for a portion (e.g., a block or macroblock) of an interlaced P-field, including selecting between using a same polarity or opposite polarity motion vector predictor for the portion. The encoder/decoder processes the motion vector based at least in part on the motion vector predictor computed for the motion vector. The processing can comprise computing a motion vector differential between the motion vector and the motion vector predictor during encoding and reconstructing the motion vector from a motion vector differential and the motion vector predictor during decoding. The selecting can be based at least in part on a count of opposite polarity motion vectors for a neighborhood around the portion and/or a count of same polarity motion vectors.

The invention discloses a remote sensing image semantic segmentation method based on attention multi-scale feature fusion, and the method comprises the steps: establishing a semantic segmentation network based on attention multi-scale feature fusion, building a training data set, and carrying out the network parameter training through employing the training data set. And performing semantic segmentation on to-be-tested data by using the trained network during testing. And the network is a lightweight encoder-decoder structure. Wherein the idea of an image cascade network is introduced; meanwhile, coding features and decoding features are optimized by using an attention mechanism, a multi-scale attention optimization module, a multi-scale feature fusion module and a boundary enhancement module are constructed, feature maps of different scales are extracted and fused, training is guided by using multi-scale semantic tags and boundary tags, and semantic segmentation of remote sensing images can be effectively carried out.

Deblocking filter used in a video encoder/decoder based on a multilayer. In deciding a deblocking filter strength when performing a deblocking filtering with respect to a boundary between a current block coded by an intra-BL mode and its neighboring block, it is determined whether the current block or the neighboring block has coefficients. The filter strength is decided as a first filter strength if it is determined that the current block or the neighboring block has the coefficients, and the filter strength is decided as a second filter strength if it is determined that the current block or the neighboring block does not have the coefficients. The first filter strength is greater than the second filter strength.

Apparatus and method to decode video data while maintaining a target video quality using an integrated error control system including error detection, resynchronization and error recovery are described. Robust error control can be provided by a joint encoder-decoder functionality including multiple error resilience designs. In one aspect, error recovery may be an end-to-end integrated multi-layer error detection, resynchronization and recovery mechanism designed to achieve reliable error detection and error localization. The error recovery system may include cross-layer interaction of error detection, resynchronization and error recovery subsystems. In another aspect, error handling of a scalable coded bitstream is coordinated across a base-layer and enhancement layer of scalable compressed video.

A prediction encoder/decoder, a prediction encoding/decoding method, and a computer readable recording medium having a program for the prediction encoding/decoding method recorded thereon. The prediction encoder includes a prediction encoding unit that starts prediction at an origin macroblock of an area of interest of a video frame, continues prediction in an outward spiral in the shape of square rings composed of macroblocks surrounding the origin macroblock, and encodes video by performing intra-prediction using information about a macroblock that has just been coded in a square ring including a macroblock to be coded and a macroblock in a previous square ring and adjacent to the macroblock.

An autonomous and portable smartcard reader device incorporates a high level of embedded security countermeasures. Data transfers are encrypted with two specific input devices, namely a light sensor and PIN or other keyboard entry, and at the output through the use of a dual-tone encoder-decoder. The unit may be used alone or as a plug-in to another device such as a PDA, cell phone, or remote control. The reader may further be coupled to various biometric or plug-in devices to achieve at least five levels of authentication, namely, (1) the smartcard itself; (2) the smartcard reader; (2) the PIN; (3) private-key cryptography (PKI); and (5) the (optional) biometric device. These five levels account for an extremely strong authentication applicable to public networking on public/private computers, and even on TV (satellite, cable, DVD, CD AUDIO, software applications. Transactions including payments may be carried out without any risk of communication tampering, authentication misconduct or identity theft. The invention finds utility in a wide range of applications, including commercial transactions and security, including the generation of prepaid credit/debit card numbers, vehicle rental, and other uses.

A bluetooth headset for transmitting and receiving digital data and analog audio signal is disclosed. The bluetooth headset comprises: a bluetooth module for transmitting and receiving a wireless signal comprising a control signal, the format of the wireless signal being compliant with a bluetooth specification; an audio encoder/decoder for encoding an analog audio input signal to generate digital data, and decoding the digital data to generate an analog audio output signal; a memory unit for storing the digital data; and a control unit which is coupled to the bluetooth module for converting the wireless signal from received by the bluetooth module into the digital data, and transferring the digital data to one of the memory unit or the encoder/decoder.

The invention discloses a deep learning-based blog text abstract generation method. The method comprises the following steps of: crawling blog data; preprocessing the crawled blog data and selecting blog text data; converting the selected blog text data into vector matrix data according to a Chinese word vector dictionary; constructing a deep learning encoder-decoder model, separately training an encoder and a decoder of the model, and connecting the encoder and the decoder for use after the training is completed; and repeating the steps S01 to S03 to obtain generated data, and generating a predicted abstract from the generated data through the trained model. According to the method, text abstracts of blogs are automatically generated on the basis of a deep learning frame encoder-decoder, and deeper semantic relation of the blogs can be obtained at the same time. The generated text abstracts can visually display the main content of the current blog, so that the text abstracts have wide application prospect.

In one aspect, an encoder/decoder receives information for four field motion vectors for a macroblock in an interlaced frame-coded, forward-predicted picture and processes the macroblock using the four field motion vectors. In another aspect, an encoder/decoder determines a number of valid candidate motion vectors and calculates a field motion vector predictor. The encoder/decoder does not perform a median operation on the valid candidates if there are less than three of them. In another aspect, an encoder/decoder determines valid candidates, determines field polarities for the valid candidates, and calculates a motion vector predictor based on the field polarities. In another aspect, an encoder/decoder determines one or more valid candidates, determines a field polarity for each individual valid candidate, allocates each individual valid candidate to one of two sets (e.g., opposite polarity and same polarity sets) depending on its field polarity, and calculates a motion vector predictor based on the two sets.

A remote interface network in which multiple remote HID encoder/decoder units share a common physical transport medium for connecting to one or more processing unit encoder/decoders is described. In one embodiment, the physical transport medium includes an optical shared media transport network. Each remote HID encoder/decoder unit can support one or more remote HIDs. The processing unit encoder/decoder can support one or more Pus. The network can be used, for example, in office, hospital, dense seat (e.g., aircraft, bus, etc.) and content provider networks.

In a multi-channel encoder generating several different parameter sets for reconstructing a multi-channel output signal using at least one transmission channel, the data stream is written such that the two parameter sets are decodable independently of each other. Thus, a multi-channel decoder is enabled to skip a parameter set which is marked as optional and/or has a higher version number when reading the data stream and still to perform a valid multi-channel reconstruction using a data set marked as mandatory or a data set having a sufficiently low version number. This achieves a flexible encoder/decoder concept suitable for future updates characterized by backward compatibility and reliability.

An encoder/decoder is based on a combination of two audio or video channels to obtain a first combination signal as a mid-signal and a residual signal derivable using a predicted side signal derived from the mid-signal. A decoder uses the prediction residual signal, the first combination signal, a prediction direction indicator and prediction information to derive decoded first channel and second channel signals. A real-to-imaginary transform can be applied for estimating the imaginary part of the spectrum of the first combination signal. The prediction signal used in the derivation of the prediction residual signal, the real-valued first combination signal is multiplied by a real portion of the complex prediction information and the estimated imaginary part of the first combination signal is multiplied by an imaginary portion of the complex prediction information.

A broadband network system configured to transfer user datagram protocol with value-added (UDPVA) packet from to a broadband interface unit (BIntU) transceiver. The broadband network system may comprise a data distribution center and the BIntU transceiver. The BIntU transceiver is configured to generate the UDPVA packet. The BIntU transceiver includes an encoder/decoder (codec) configured to alternatively code or decode UDP frame information and a digital signal processor (DSP) portion. The DSP portion is coupled to the codec. In one aspect, the DSP portion includes a stack that temporarily stores the UDPVA packet within the stack. In one aspect, the UDPVA packet is in a form to be transmitted directly to a network destination address device. In another aspect, the UDPVA packet is available to delivery to a network destination address or storage located on a local area network or a wide area network. In certain aspects, the BIntU transceiver is configured to transmit the UDPVA packet to the data distribution center.

The invention discloses an aspect-level emotion classification model and method based on dual-memory attention, belonging to the technical field of text emotion classification. The model of the invention mainly comprises three modules: an encoder composed of a standard GRU loop neural network, a GRU loop neural network decoder introducing a feedforward neural network attention layer and a Softmaxclassifier. The model treats input statements as a sequence, based on the attention paid to the position of the aspect-level words in the sentence, Two memory modules are constructed from the originaltext sequence and the hidden layer state of the encoder respectively. The randomly initialized attention distribution is fine-tuned through the attention layer of the feedforward neural network to capture the important emotional features in the sentences, and the encoder-decoder classification model is established based on the learning ability of the GRU loop neural network to the sequence to achieve aspect-level affective classification capabilities. The invention can remarkably improve the robustness of the text emotion classification and improve the classification accuracy.

The invention discloses a multi-feature cyclic convolution significance target detection method based on an attention mechanism. The method comprises the following steps: ; the method comprises the following steps of: 1, analyzing common characteristics of a salient target in a natural image, including spatial distribution and contrast characteristics, using an improved U-Net full convolutional neural network, performing pixel-by-pixel prediction by adopting an encoder-decoder structure, and performing multi-level and multi-scale characteristic fusion between an encoder and a decoder by adopting a cross-layer connection mode; secondly, a large number of clutters can be introduced to interfere with the generation of a final prediction graph by carrying out concentage fusion on coding end features and decoding end features, so that an attention module is introduced to calibrate full-pixel weights from two angles between channels and between pixels, the task-related pixel weights are enhanced, and the background and noise influence is weakened; and 3, a multi-feature cyclic convolution module is used as a post-processing means, the spatial resolution capability is enhanced through iteration, the edge of an image region is further refined and segmented, and a finer significant target mask is obtained.

A lightweight semantic segmentation method for a high-resolution remote sensing image comprises the steps of network construction, training and testing. Specifically, a deep semantic segmentation network of an encoder-decoder structure is constructed for a pytorch deep learning framework, after network training is carried out based on a remote sensing image data sample set, a to-be-tested remote sensing image serves as network input. A segmentation result of the remote sensing image is obtained. According to the method, on one hand, model parameters are reduced by decomposing depth separable convolution, the calculation complexity is reduced, the semantic segmentation time of the high-resolution remote sensing image is shortened, and the semantic segmentation efficiency of the high-resolution remote sensing image is improved; and on the other hand, semantic segmentation precision is improved through multi-scale feature aggregation, a spatial attention module and gating convolution, sothat the proposed lightweight deep semantic segmentation network can accurately and efficiently realize semantic segmentation of a high-resolution remote sensing image.

A method, medium, and system compressing and/or reconstructing image data. The data compression method includes compressing current data according to a plurality of modes for compressing current data, determining whether the current data compressed according to the plurality of modes can be represented by a fixed length of bits, selecting a mode from among modes in which the compressed current data can be represented by the fixed length of bits, and outputting the compressed current data according to the selected mode. Accordingly, it is possible to significantly lower the complexity of an image encoder/decoder system, and exactly meet a picture-based Control Bit Rate (CBR) suitable for a Liquid Crystal Display Dynamic Capacitance Compensation (LCD DCC) device/system.

A method and apparatus for encoding/decoding information using low density parity check (LDPC) codewords of varying lengths. In one implementation, an encoder/decoder uses a parity check matrix having a size that corresponds to a length of information for each codeword. In one example, the size of the parity check matrix may be retrieved from multiple stored matrices having different sizes. In other examples, a mother code matrix may be dynamically adjusted by puncturing parity bits and/or deleting information bits. Other inventive embodiments and adaptations are also disclosed.