A5025863422- Human Motion and Animation
- Human Pose and Action Recognition
- Hand Gesture Recognition Systems
- Generative Adversarial Networks and Image Synthesis
- Colorectal Cancer Screening and Detection
- Internet Traffic Analysis and Secure E-voting
- Network Traffic and Congestion Control
- Music and Audio Processing
- Software-Defined Networks and 5G
- Urinary and Genital Oncology Studies
- Hearing Impairment and Communication
- Retinoids in leukemia and cellular processes
- Bladder and Urothelial Cancer Treatments
- Image Enhancement Techniques
- Cell death mechanisms and regulation
- Speech and dialogue systems
- Gait Recognition and Analysis
- Neuroblastoma Research and Treatments
- Advanced Image Processing Techniques
- Recommender Systems and Techniques
- Technology and Data Analysis
- Advanced Vision and Imaging
Guangzhou Medical University
2021-2025
Tsinghua University
2019-2023
First Affiliated Hospital of Fujian Medical University
2022
Fujian Medical University
2022
Despite many years of improvements to it, TCP still suffers from an unsatisfactory performance. For services dominated by short flows (e.g., web search and e-commerce), the flow startup problem cannot fully utilize available bandwidth in modern Internet: starts a conservative static initial window (IW, 2-4 or 10), while most are too converge best sending rate before session ends. long video streaming file downloading), congestion control (CC) scheme manually statically configured might not...
Dance and music are two highly correlated artistic forms. Synthesizing dance motions has attracted much attention recently. Most previous works conduct music-to-dance synthesis via directly to human skeleton keypoints mapping. Meanwhile, choreographers design from in a two-stage manner: they firstly devise multiple choreographic units (CAUs), each with series of motions, then arrange the CAU sequence according rhythm, melody emotion music. Inspired by these, we systematically study such...
Abstract Neuroblastoma (NB) is a common malignant and solid pediatric tumor with unfavorable prognosis. Although studies have shown the anti-tumor efficacy of lentinan (LNT), molecular mechanism that contribute to effect on NB remains unclear. The aim this study unmask role LNT in specific mechanism. At first, vivo experiments were conducted results indicated could suppress growth NB. Subsequent cellular functional assays unveiled treatment efficiently decrease cell viability, induce cycle...
Identifying disease burden and risk factors of bladder cancer projecting its epidemiological trend in China, which can provide reference data to formulate measures for management prevention.We analyzed the incidence, mortality, disability-adjusted life-years (DALYs) China from 1990 2019 predicted 2030 based on Global Burden Disease Study 2019. We also estimated proportion contributing DALYs. The average annual percentage change (AAPC) both sexes was calculated quantify temporal trends.In...
Neuroblastoma (NB) is a solid tumor in the nervous system and has high mortality rate children. Curcumin well-characterized anticancer properties, while there no effective method clinical treatment. MTT assays revealed that curcumin dramatically inhibited proliferation of SK-N-SH cells. Compared with control group, markedly restrained migration induced cell apoptosis by G2/M cycle arrest activated caspase-3 activity. Furthermore, promoted overproduction intracellular ROS activating p53 Bcl-2...
The aim of human motion modeling is to understand behaviors and create reasonable like real people given different priors. With the development deep learning, researchers tend leverage data-driven methods improve performance traditional methods. In this paper, we present a comprehensive survey recent researches. We discuss three categories researches: prediction, humanoid control cross-modal synthesis provide detailed review over existing Finally, further remaining challenges in modeling.
Human hands, the primary means of non-verbal communication, convey intricate semantics in various scenarios. Due to high sensitivity individuals hand motions, even minor errors motions can significantly impact user experience. Real applications often involve multiple avatars with varying shapes, highlighting importance maintaining across avatars. Therefore, this paper aims transfer motion between diverse based on their respective models. To address problem, we introduce a novel anatomy-based...
Sentiment plays an essential role in people's perception of images. To incorporate the sentiment information into image style transfer task for better sentiment-aware performance, we introduce a new named transfer. solve this problem, first novel Multi-Sentiment Semantics Space (MSS-Space) to capture non-deterministic and complicated nature semantics. With MSS-Space, establish tight associations between visual attributes images multi-sentiment semantics by minimizing their distance MSS-Space...
Capturing and maintaining geometric interactions among different body parts is crucial for successful motion retargeting in skinned characters. Existing approaches often overlook geometries or add a geometry correction stage after skeletal retargeting. This results conflicts between skeleton interaction correction, leading to issues such as jittery, interpenetration, contact mismatches. To address these challenges, we introduce new framework, MeshRet, which directly models the dense...
There are anecdotal evidences that realtime mobile multiplayer games (RMMGs), which require interactions, suffer from unsatisfying experience in the wild. This paper presents thefirst measurement results of such based on 12 million real game sessions a toptier RMMG. We observe 13% at least one location resynchronization, and 7.12% have been aborted abnormally before end game. thus proposes ExCause, general causal analysis framework to systematically analyze historical session records 1)...
Synthesizing co-speech gestures is challenging because the mapping from speech to gesticulation inherently non-deterministic. When giving talks, people conduct not only gentle and rhythmic motions but also abrupt salient gesticulations. Most previous research efforts, however, ignore this nature of synthesize deterministic results, producing over-smoothed movements with limited expressiveness. To address issue, we propose a new gesture generation approach that produces high-quality...