• Journal of the Korea Convergence Society
• /
• v.13 no.1
• /
• pp.13-21
• /
• 2022

Recent deep learning-based research shows excellent performance in most natural language processing (NLP) fields with pre-trained language models. In particular, the auto-encoder-based language model proves its excellent performance and usefulness in various fields of Korean language understanding. However, the decoder-based Korean generative model even suffers from generating simple sentences. Also, there is few detailed research and data for the field of conversation where generative models are most commonly utilized. Therefore, this paper constructs multi-turn dialogue data for a Korean generative model. In addition, we compare and analyze the performance by improving the dialogue ability of the generative model through transfer learning. In addition, we propose a method of supplementing the insufficient dialogue generation ability of the model by extracting recommended response candidates from external knowledge information through a retrival model.

• Tunnel and Underground Space
• /
• v.29 no.3
• /
• pp.184-196
• /
• 2019

This study aims to extract a 3D image of micro-cracks generated by hydraulic fracturing tests, using the deep learning method and X-ray computed tomography images. The pixel-level cracks are difficult to be detected via conventional image processing methods, such as global thresholding, canny edge detection, and the region growing method. Thus, the convolutional neural network-based encoder-decoder network is adapted to extract and analyze the micro-crack quantitatively. The number of training data can be acquired by dividing, rotating, and flipping images and the optimum combination for the image augmentation method is verified. Application of the optimal image augmentation method shows enhanced performance for not only the validation dataset but also the test dataset. In addition, the influence of the original number of training data to the performance of the deep learning-based neural network is confirmed, and it leads to succeed the pixel-level crack detection.

• The KIPS Transactions:PartA
• /
• v.11A no.1
• /
• pp.81-88
• /
• 2004

In this paper, we designed the outer encoder/decoder for the terrestrial DMB that is an advanced digital broadcasting standard, implemented, and verified by using ALTERA FPGA. In the encoder part, it was created the parity bytes (16 bytes) from the input packet (188by1e) of MPEG-2 TS and the encoded data was distributed output by the convolutional interleaver for Preventing burst errors. In the decoder part, It was proposed the algorithm that detects synchronous character suitable to DMB in transmitted data from the encoder. The circuit complexity in RS decoder was reduced by applying a modified Euclid's algorithm. This system has a capability to correct error of the maximum 8 bytes in a packet. After the outer encoder/decoder algorithm was verified by using C language, described in VHDL and implemented in the ALTERA FPGA chips.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.39 no.4
• /
• pp.104-111
• /
• 2002

In this paper, we proposed the architecture of the decoder which implements the multi-view images decoding algorithm. The study of the hardware structure of the multi-view image processing has not been accomplished. The proposed multi-view images decoder operates in a three stage pipelined manner and extracts the depth of the pixels of the decoded image every clock. The multi-view images decoder consists of three modules, Node selector which transfers the value of the nodes repeatedly and Depth Extractor which extracts the depth of each pixel from the four values of the nodes and Affine transformer which generates the projecting position on the image plane from the values of the pixels and the specified viewpoint. The proposed architecture is designed and simulated by the Max+plus II design tool and the operating frequency is 30MHz. The image can be constructed in a real time by the decoder with the proposed architecture.

• Convergence Security Journal
• /
• v.18 no.5_1
• /
• pp.19-24
• /
• 2018

For a long time now, general-purpose processors have provided dedicated hardware / software tracing modules to provide developers with tools to fix bugs. A hardware tracer generates its enormous data into a log that is used for both performance analysis and debugging. Processor Trace (PT) is a new hardware-based tracing feature for Intel CPUs that traces branches executing on the CPU, which allows the reconstruction of the control flow of all executed code with minimal labor. Hardware tracer has been integrated into the operating system, which allows tight integration with its profiling and debugging mechanisms. However, in the Windows environment, existing studies related to PT focused on decoding only one flow in sequence. In this paper, we propose an extended PT decoder structure that provides basic data for real-time trace and malicious code detection using the functions provided by PT in Windows environment.

• The Journal of the Acoustical Society of Korea
• /
• v.42 no.2
• /
• pp.94-101
• /
• 2023

A deep neural network composed of encoders and decoders, such as U-Net, used for speech enhancement, concatenates the encoder to the decoder through skip-connection. Skip-connection helps reconstruct the enhanced spectrum and complement the lost information. The features of the encoder and the decoder connected by the skip-connection are incompatible with each other. In this paper, for complex-valued spectrum based speech enhancement, Self-Attention (SA) method is applied to skip-connection to transform the feature of encoder to be compatible with the features of decoder. SA is a technique in which when generating an output sequence in a sequence-to-sequence tasks the weighted average of input is used to put attention on subsets of input, showing that noise can be effectively eliminated by being applied in speech enhancement. The three models using encoder and decoder features to apply SA to skip-connection are studied. As experimental results using TIMIT database, the proposed methods show improvements in all evaluation metrics compared to the Deep Complex U-Net (DCUNET) with skip-connection only.

• Journal of Broadcast Engineering
• /
• v.11 no.1 s.30
• /
• pp.116-129
• /
• 2006

The progress of data transmission technology through the Internet has spread a variety of realistic contents. One of such contents is multi-view video that is acquired from multiple camera sensors. In general, the multi-view video processing requires encoders and decoders as many as the number of cameras, and thus the processing complexity results in difficulties of practical implementation. To solve for this problem, this paper considers a simple multi-view system utilizing a single encoder and a single decoder. In the encoder side, input multi-view YUV sequences are combined on GOP units by a video mixer. Then, the mixed sequence is compressed by a single H.264/AVC encoder. The decoding is composed of a single decoder and a scheduler controling the decoding process. The goal of the scheduler is to assign approximately identical number of decoded frames to each view sequence by estimating the decoder utilization of a Gap and subsequently applying frame skip algorithms. Furthermore, in the frame skip, efficient frame selection algorithms are studied for H.264/AVC baseline and main profiles based upon a cost function that is related to perceived video quality. Our proposed method has been performed on various multi-view test sequences adopted by MPEG 3DAV. Experimental results show that approximately identical decoder utilization is achieved for each view sequence so that each view sequence is fairly displayed. As well, the performance of the proposed method is examined in terms of bit-rate and PSNR using a rate-distortion curve.

• The Transactions of the Korea Information Processing Society
• /
• v.7 no.11S
• /
• pp.3668-3674
• /
• 2000

Recently Technology regarding VoIP (Voice over IP) is emerging over the market of the IP network. So far nothing is unfortunately there any attempt to try any research with respect to the development of the protocol stack relating to such control of gateway as MGCP, MEGACO, SIP, SDP. The reasons come from the low level of infrastructue, the shortage of the time and technology required at the moment, and so on. In this regards, this paper is focused on developing a protocol stack made with encoder/decoder, the generator of the header file etc, based on the protocol grammars of MGCP, SDP supported by IETF. For the sake of it, first develops the syntax analyzer, encoder/decoder, header file generator for encoding/decoding as applying the method of syntax-directed to each protocol grammar.

• Annual Conference on Human and Language Technology
• /
• 2021.10a
• /
• pp.391-395
• /
• 2021

텍스트 스타일 변환은 입력 스타일(source style)로 쓰여진 텍스트의 내용(content)을 유지하며 목적 스타일(target style)의 텍스트로 변환하는 문제이다. 텍스트 스타일 변환을 시퀀스 간 변환 문제(sequence-to-sequence)로 보고 기존 기계학습 모델을 이용해 해결할 수 있지만, 모델 학습에 필요한 각 스타일에 대응되는 병렬 말뭉치를 구하기 어려운 문제점이 있다. 따라서 최근에는 비병렬 말뭉치를 이용해 텍스트 스타일 변환을 수행하는 방법들이 연구되고 있다. 이 연구들은 주로 인코더-디코더 구조의 생성 모델을 사용하기 때문에 입력 문장이 가지고 있는 내용이 누락되거나 다른 내용의 문장이 생성될 수 있는 문제점이 있다. 본 논문에서는 마스크 언어 모델(masked language model)을 이용해 입력 텍스트의 내용을 유지하면서 원하는 스타일로 변경할 수 있는 텍스트 스타일 변환 방법을 제안하고 한국어 긍정-부정, 채팅체-문어체 변환에 적용한다.

• Proceedings of the IEEK Conference
• /
• 2001.09a
• /
• pp.477-480
• /
• 2001

본 논문에서는 허프만 코딩의 기본 원리인 이진트리로부터 생성된 효율적인 일차원 검색테이블 구성과, 프로세싱 효율을 증가시키기 위한 수치연산 기법의 도입을 통하여 도출된 효율적인 허프만 디코딩 방법을 제시한다. 제안된 방법에서는 허프만 코드화 된 비트스트림으로부터 읽어 들인 비트 데이터를 직접 이진트리를 검색하기 위한 주소연산에 적용함으로써, 기존의 이진트리 검색을 위하여 수행되던 비교 및 분기문의 제거를 가능토록 하였다. 본 논문에서 제안하는 허프만 디코딩 기법은 기존의 비교 및 분기구분에 의하여 수행되던 허프만 디코딩 기법대비 30%이상의 검색속도 증가 및 이진트리 방식의 검색을 수행하기 위하여 필요로 하는 검색테이블에서 존재하던 데이터 및 노드간의 연결주소를 저장하기 위하여 사용되던 메모리 공간을 효율적으로 제거할 수 있는 일차원 검색테이블을 구성함으최써, 기존 방법대비 1/3의 메모리 공간만을 사용하도록 하였다. 제안된 허프만 디코딩 기법의 성능평가를 위하여 MPEG-2 AAC의 허프만 디코더에 적용한 결과 기존의 순차검색방법에 의한 디코딩 기법대비 198%의 성능향상 결과를 확인할 수 있었다.