The ability to animate photo-realistic head avatars reconstructed from monocular portrait video sequences represents a crucial step in bridging the gap between virtual and real worlds. Recent advancements avatar techniques, including explicit 3D morphable meshes (3DMM), point clouds, neural implicit representation have been exploited for this ongoing research. However, 3DMM-based methods are constrained by their fixed topologies, point-based approaches suffer heavy training burden due...

The multiple-input-multiple-output (MIMO) and cooperative communication are two state-of-the-art techniques to provide high-rate high-quality video services. By taking advantage of both techniques, this paper presents a novel joint source-channel coding (JSCC) framework for scalable transmission over MIMO. In framework, we first propose MIMO architecture, which employs macro-micro power control strategy as relaying protocol determine the on/off mode relays specific allocation among them...

Three-dimensional (3-D) video applications, such as 3-D cinema, 3DTV, and Free Viewpoint Video (FVV) are attracting more attention both from the industry in literature. High accuracy of depth is a fundamental prerequisite for most applications. However, accurate requires computationally intensive global optimization. This high computational complexity one bottlenecks to applying generation especially mobile networks since terminals usually have limited computing ability. paper presents...

This paper investigates the issue of delay-constrained video delivery over IEEE 802.11 networks. We analyze characteristics MAC from viewpoint queuing theory and packet loss probability for by jointly considering forward error correction (FEC) decoding deadline. By taking efficient shortened-and-punctured Reed-Solomon codes as basis application-FEC, we propose a congestion-adaptive packets scheduling (CPS) scheme to add unequal protection (UEP) function into 802.11-based system transmission....

The ability to animate photo-realistic head avatars reconstructed from monocular portrait video sequences represents a crucial step in bridging the gap between virtual and real worlds. Recent advancements avatar techniques, including explicit 3D morphable meshes (3DMM), point clouds, neural implicit representation have been exploited for this ongoing research. However, 3DMM-based methods are constrained by their fixed topologies, point-based approaches suffer heavy training burden due...

This paper introduces Stereo-Talker, a novel one-shot audio-driven human video synthesis system that generates 3D talking videos with precise lip synchronization, expressive body gestures, temporally consistent photo-realistic quality, and continuous viewpoint control. The process follows two-stage approach. In the first stage, maps audio input to high-fidelity motion sequences, encompassing upper-body gestures facial expressions. To enrich diversity authenticity, large language model (LLM)...

In this paper, we introduce a multi-agent simulation framework Casevo (Cognitive Agents and Social Evolution Simulator), that integrates large language models (LLMs) to simulate complex social phenomena decision-making processes. is designed as discrete-event simulator driven by agents with features such Chain of Thoughts (CoT), Retrieval-Augmented Generation (RAG), Customizable Memory Mechanism. enables dynamic modeling, which can support various scenarios network analysis, public opinion...

e20526 Background: Organoids derived from cancer patients are considered to represent the overall appearance of original tumor in lung cancer. With in-depth study molecular mechanisms and development new drugs, survival time advanced non-small cell has been greatly extended. However, when it comes back-end treatment, they often have limited amounts fresh tissue samples or cannot complete puncture, which affects successfully establishment organoids. Some malignant pleural effusion (MPE)...

Guided by the concept of next generation broadcasting, China's broadcasting network is undergoing a migration from DVB based system to an all IP-based one so that triple-play services can be delivered enhance competitiveness Chinese service providers. This change happens primarily backbone networks, which are IP/MPLS based. The integrated will introduce multicast traffics require isolated delivery with high availability and robustness for safety, control management. Thus, implementation...

For resource-constrained wireless scenarios, this paper proposes a cooperative cross-layer scheme to improve the performance of scalable video transmission. At application layer, extraction parameters coding are first optimized maximize visual quality and fulfill resource constraints decoding capacity bandwidth. link priority-based scheduling strategy is further proposed transmit packets extracted substream. In strategy, relay may substitute for unreliable direct if it has relatively smaller...

This paper proposes a cooperative multiple-input multiple-output (MIMO) architecture to transmit the video reliably. By exploring configurable resources of scalable and coded MIMO system, we further propose novel joint source-channel coding (JSCC) framework with optimal tradeoff among diversity, multiplexing gains. In this framework, concatenated low-density parity-check codes (LDPC) diversity-embedded space-time block (DE-STBC) provide double unequal error protection for layers, STBCs...

Home wireless monitoring problem is the subject of today's cutting-edge research in field International Intelligent Building.Advent sensor networks to overcome household wiring cumbersome, fully embodies flexibility smart home system, convenient and efficient.The project was developed based on system ZigBee technology hitemet technology, remote control short distance combined achieve data acquisition systems, avoiding cabling, scalability.Firstly, overall design with underlying network...

Bit-depth scalability is a new research field in the on-going scalable extension of H.264/AVC (SVC) video coding standard. The key point to accurately predict enhancement layer, whose bit depth 10 or more, from 8 base layer. An improved inter-layer prediction scheme for bit-depth was developed that ensures compatibility with standard and improves encoding efficiency. uses “perfect” picture an adaptive neighbor filter coefficients are optimized by minimizing block distortion between...

An opportunistic automatic repeat request (ARQ) scheme was developed to support high-quality video transmission over cooperative decode-forward networks. The ARQ is implemented in the form of distributed space-time block codes (STBCs) and desired STBC for a certain channel state determined by diversity-multiplexing tradeoff principle. With good states, source relay may jointly retransmit lost packets using high-rate low-diversity while with bad states they will low-rate high-diversity codes....