• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2011년도 하계학술대회
• /
• pp.352-355
• /
• 2011

본 논문에서는 깊이 맵의 재배열 과정을 통해서, 보다 개선된 영상을 합성하는 방법을 제안한다. 제안하는 방법은 전체 깊이 맵을 여러 그룹(Group)으로 나누고, 각각의 그룹에 서로 다른 가중치를 주어 가까운 물체에 좀 더 많은 깊이 값을 가질수 있도록 조절하였다. 깊이 맵 추정(Depth Estimation) 및 중간 시점 영상의 합성(View Synthesis)을 통하여 기존 방식과의 비교를 진행하였고 그 결과, 전체적인 비디오 시퀀스(Video Sequence)에 대한 PSNR은 유지하면서, 보다 시각적으로 자연스러운 영상을 얻을 수 있었다.

• 전자공학회논문지
• /
• 제51권3호
• /
• pp.139-143
• /
• 2014

This paper proposes a simplified DC calculation method for simplified depth coding (SDC) mode of 3D High Efficiency Video Coding (3D-HEVC) to reduce the computational complexity. For the computational complexity reduction, the current reference software of 3D-HEVC employs reference samples sub-sampling method. However, accumulation, branch, and division operations are still utilized and these operations increase computational complexity. The proposed method calculates DC value without those operations. The experimental results show that the proposed method achieves 0.1% coding gain for synthesized views in common test condition (CTC) with the significantly reduced number of computing operations.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2012년도 하계학술대회
• /
• pp.31-33
• /
• 2012

MPEG 3DV 그룹에서는 재생되는 시점 수 보다 적은 시점의 다시점 비디오와 그에 대응하는 깊이영상 및 관련 파라미터를 하나의 비트스트림으로 부호화하는 다시점 및 깊이영상 부호화(MVD: Multiview Video plus Depth) 표준화가 진행 중에 있다. 본 논문은 3DV MVD 비트스트림 포맷을 MPEG-2 TS 로 다중화하기 위한 TS 다중화 확장기법을 제안한다. 또한, MVD 재현 시나리오에 따른 효율적인 TS 전송 구조를 제시하고, TS 다중화 SW 툴의 설계 및 구현을 통하여 그 장단점을 고찰한다. 제안된 기법은 프로그램 및 ES 에 대한 정보를 함께 다중화하여 MVD 의 다양한 재현 시나리오 적응 및 효율적인 복호화 및 가상시점 합성을 지원할 수 있다.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송∙미디어공학회 2020년도 하계학술대회
• /
• pp.295-296
• /
• 2020

3DoF+ 비디오 부호화 표준을 개발하고 있는 MPEG-I 비주얼 그룹은 표준화 과정에서 참조 SW 코덱인 TMIV(Test Model for Immersive Video)를 개발하고 있다. TMIV 는 제한된 공간에서 동시에 여러 위치에서 획득한 뷰(view)의 텍스처(texture) 비디오와 깊이(depth) 비디오를 효율적으로 압축하여 임의 시점의 뷰 렌더링(rendering)을 제공한다. TMIV 에서 수행되는 깊이 비디오의 비트 심도 스케일링 및 압축은 깊이 정보의 손실을 발생하며 이는 렌더링(rendering)된 임의 시점 비디오의 화질 저하를 야기한다. 본 논문에서는 보다 효율적인 깊이 비디오 압축을 위한 히스토그램 등화(histogram equalization) 기반의 구간별(piece-wise) 깊이 매핑 기법을 제안한다. 실험결과 제안기법은 자연 영상(natural sequence)의 End-to-End 부호화 성능에서 평균적으로 3.1%의 비트율 절감이 있음을 확인하였다.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2013년도 하계학술대회
• /
• pp.182-184
• /
• 2013

A HEVC-based scalable 3D video coding system is proposed. The proposed system supports scalable transmission of multiview video data with depth maps. Key technologies in this system are reference picture management, reference picture list construction, and cross-layer dependency signaling. All the proposed technologies are used for the development of video coding system for UHD stereo display and glassless 3D display.

• IEIE Transactions on Smart Processing and Computing
• /
• 제5권4호
• /
• pp.231-234
• /
• 2016

The next-generation video coding standard known as High-Efficiency Video Coding (HEVC) was developed with the aim of doubling the bitrate reduction offered by H.264/Advanced Video Coding (AVC) at the expense of an increase in computational complexity. Mode decision with motion estimation is still one of the most time-consuming computations in HEVC, as it is with H.264/AVC. Several schemes for a fast mode decision have been presented in reference software and in other studies. However, a possible speed-up in conventional schemes is sometimes insignificant for videos that have inhomogeneous spatial and temporal characteristics. This paper proposes a bypass algorithm to skip large-block-size predictions for videos where small block sizes are preferred over large ones. The proposed algorithm does not overlap with those in previous works, and thus, is easily used with other fast algorithms. Consequently, an independent speed-up is possible.

• 한국방송∙미디어공학회:학술대회논문집
• /
• 한국방송공학회 2011년도 추계학술대회
• /
• pp.288-291
• /
• 2011

Video stabilization is an important image enhancement widely used in surveillance system in order to improve recognition performance. Most previous methods calculate inter-frame homography to estimate global motion. These methods are relatively slow and suffer from significant depth variations or multiple moving object. In this paper, we propose a fast and practical approach for video stabilization that selects the most reliable key frame as a reference frame to a current frame. We use optical flow to estimate global motion within an adaptively selected region of interest in static camera environment. Optimal global motion is found by probabilistic voting in the space of optical flow. Experiments show that our method can perform real-time video stabilization validated by stabilized images and remarkable reduction of mean color difference between stabilized frames.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• 제16권7호
• /
• pp.2407-2424
• /
• 2022

Recently, Transformer has made great progress in video retrieval tasks due to its high representation capability. For the structure of a Transformer, the cascaded self-attention modules are capable of capturing long-distance feature dependencies. However, the local feature details are likely to have deteriorated. In addition, increasing the depth of the structure is likely to produce learning bias in the learned features. In this paper, an improved Transformer structure named TransDCS (Transformer with Dynamic Convolution and Shortcut) is proposed. A Multi-head Conv-Self-Attention module is introduced to model the local dependencies and improve the efficiency of local features extraction. Meanwhile, the augmented shortcuts module based on a dual identity matrix is applied to enhance the conduction of input features, and mitigate the learning bias. The proposed model is tested on MSRVTT, LSMDC and Activity-Net benchmarks, and it surpasses all previous solutions for the video-text retrieval task. For example, on the LSMDC benchmark, a gain of about 2.3% MdR and 6.1% MnR is obtained over recently proposed multimodal-based methods.

• Journal of Dental Anesthesia and Pain Medicine
• /
• 제24권2호
• /
• pp.81-90
• /
• 2024

Nasotracheal intubation (NTI) plays an important role in pediatric airway management, offering advantages in specific situations, such as oral and maxillofacial surgery and situations requiring stable tube positioning. However, compared to adults, NTI in children presents unique challenges owing to anatomical differences and limited space. This limited space, in combination with a large tongue and short mandible, along with large tonsils and adenoids, can complicate intubation. Owing to the short tracheal length in pediatric patients, it is crucial to place the tube at the correct depth to prevent it from being displaced due to neck movements, and causing injury to the glottis. The equipment used for NTI includes different tube types, direct laryngoscopy vs. video laryngoscopy, and fiberoptic bronchoscopy. Considering pediatric anatomy, the advantages of video laryngoscopy have been questioned. Studies comparing different techniques have provided insights into their efficacy. Determining the appropriate size and depth of nasotracheal tubes for pediatric patients remains a challenge. Various formulas based on age, weight, and height have been explored, including the recommendation of depth-mark-based NTI. This review provides a comprehensive overview of NTI in pediatric patients, including the relevant anatomy, equipment, clinical judgment, and possible complications.

• 한국콘텐츠학회지
• /
• 제8권1호
• /
• pp.12-17
• /
• 2010