Huifang Sun

ORCID: 0000-0003-2918-1522
Publications
Citations
Views
---
Saved
---
About
Contact & Profiles
Research Areas
  • Video Coding and Compression Technologies
  • Advanced Data Compression Techniques
  • Advanced Vision and Imaging
  • Image and Video Quality Assessment
  • Advanced Image Processing Techniques
  • Video Analysis and Summarization
  • Image and Signal Denoising Methods
  • Seaweed-derived Bioactive Compounds
  • Marine Sponges and Natural Products
  • Echinoderm biology and ecology
  • Music and Audio Processing
  • Membrane Separation Technologies
  • Water Treatment and Disinfection
  • Multimedia Communication and Technology
  • Advanced Image Fusion Techniques
  • Proteoglycans and glycosaminoglycans research
  • Digital Filter Design and Implementation
  • Image Enhancement Techniques
  • Membrane-based Ion Separation Techniques
  • Human Pose and Action Recognition
  • Image Retrieval and Classification Techniques
  • Wastewater Treatment and Nitrogen Removal
  • Medical Image Segmentation Techniques
  • Visual perception and processing mechanisms
  • Pharmaceutical and Antibiotic Environmental Impacts

Minzu University of China
2023-2025

Shanxi University
2012-2025

Peng Cheng Laboratory
2023-2025

South Central Minzu University
2022-2023

Kunming Institute of Botany
2017-2022

Chinese Academy of Sciences
2013-2022

Shandong University
2019-2022

Mitsubishi Electric (United States)
2012-2021

University of Chinese Academy of Sciences
2013-2021

Nanyang Technological University
2017-2021

Throughout this article, we concentrate on the transcoding of block-based video coding schemes that use hybrid discrete cosine transform (DCT) and motion compensation (MC). In such schemes, frames sequence are divided into macroblocks (MBs), where each MB typically consists a luminance block (e.g., size 16 × 16, or alternatively, four 8 blocks) along with corresponding chrominance blocks Cb Cr). This article emphasizes processing is done components video. general, can be handled similarly...

10.1109/msp.2003.1184336 article EN IEEE Signal Processing Magazine 2003-03-01

State-of-the-art algorithms for blind image quality assessment (BIQA) typically have two categories. The first category approaches extract natural scene statistics (NSS) as features based on the statistical regularity of images. second by feature encoding with respect to a learned codebook. However, several problems need be addressed in existing codebook-based BIQA methods. First, high-dimensional are memory-consuming and risk over-fitting. Second, there is semantic gap between constructed...

10.1109/tmm.2017.2763321 article EN IEEE Transactions on Multimedia 2017-10-25

Traditional in the wild image quality assessment (IQA) models are generally trained with labels of mean opinion score (MOS), while missing rich subjective information contained ratings, for example, standard deviation scores (SOS) or even distribution (DOS). In this paper, we propose a novel IQA method named RichIQA to explore rating beyond MOS predict wild. is characterized by two key designs: (1) three-stage prediction network which exploits powerful feature representation capability...

10.48550/arxiv.2409.05540 preprint EN arXiv (Cornell University) 2024-09-09

The idea of moving picture expert group (MPEG) bitstream scaling relates to altering or the amount data in a previously compressed MPEG bitstream. new scaled conforms constraints that are not known nor considered when original preceded was constructed. Numerous applications for video transmission and storage being developed based on coding standard. Applications such as demand, trick-play track digital tape recorders (VTR's) extended-play recording VTR's motivate scaling. In this paper, we...

10.1109/76.488826 article EN IEEE Transactions on Circuits and Systems for Video Technology 1996-04-01

In this paper, we present a novel system and effective algorithms for soccer video segmentation. The output, about whether the ball is in play, reveals high-level structure of content. first step to classify each sample frame into 3 kinds view using unique domain-specific feature, grass-area-ratio. Here grass value classification rules are learned automatically adjusted new clip. Then heuristic used processing label sequence, obtain play/break status game. results provide good basis detailed...

10.1109/icme.2001.1237822 article EN 2022 IEEE International Conference on Multimedia and Expo (ICME) 2001-01-01

In this paper, we present algorithms for parsing the structure of produced soccer programs. The problem is important in context a personalized video streaming and browsing system. While prior work focuses on detection special events such as goals or corner kicks, paper concerned with generic structural elements game. We begin by defining two mutually exclusive states game, play break based rules soccer. select domain-tuned feature set, dominant color ratio motion intensity, syntax content...

10.1109/icassp.2002.5745558 article EN IEEE International Conference on Acoustics Speech and Signal Processing 2002-05-01

This paper describes an algorithm which can achieve a constant bit rate when coding multiple video objects. The implementation is nontrivial extension of the MPEG-4 control for single objects employs quadratic quantizer model. organized into two stages: pre- and post-encoding stage. In pre-encoding stage, initial target estimate made each object. Based on buffer fullness, total adjusted then distributed proportional to relative size, motion, variance new individual targets rate-quantizer...

10.1109/76.744285 article EN IEEE Transactions on Circuits and Systems for Video Technology 1999-01-01

The linear regression model is a very attractive tool to design effective image interpolation schemes. Some regression-based algorithms have been proposed in the literature, which objective functions are optimized by ordinary least squares (OLS). However, it shown that with OLS may some undesirable properties from robustness point of view: even small amounts outliers can dramatically affect estimates. To address these issues, this paper we propose novel algorithm based on regularized local...

10.1109/tip.2011.2150234 article EN IEEE Transactions on Image Processing 2011-05-13

(1) Background: Toll-like receptor 2 (TLR2) on platelets is increasingly recognized as a pivotal mediator in infection-induced platelet activation and aggregation, contributing to both inflammatory thrombotic diseases. Targeting TLR2 offers promising therapeutic strategy for thrombotic-related disorders. However, inhibitors targeting have not yet been reported. (2) Methods: Platelet aggregation was assessed using light transmission aggregometer. evaluated by measuring the release of...

10.3390/md23030110 article EN cc-by Marine Drugs 2025-03-04

In this paper, a new interframe coding technique based upon vector quantization is presented. This algorithm has as its basis twodimensional block quantization, at the frame level, onto which grafted concept of adaptive codebook replenishment and replenishment. includes three strategies: label replenishment, with mean shift, shift selective codeword replacement. The last two strategies efficiently update to track changes in local statistics on basis. Compared three-dimensional [16], for bit...

10.1109/tcom.1986.1096600 article EN IRE Transactions on Communications Systems 1986-07-01

To understand the formation and release behavior of iron corrosion products in a drinking water distribution system, annular reactors (ARs) were used to investigate development processes biofilm community as well concomitant behavior. Results showed that transformation bacterial are closely related each other. The presence sulfate-reducing bacteria (SRB, e.g. Desulfovibrio Desulfotomaculum), sulfur-oxidizing (SOB, Sulfuricella), iron-oxidizing (IOB, Acidovorax, Gallionella, Leptothrix,...

10.1039/c3em00544e article EN Environmental Science Processes & Impacts 2014-01-01

10.1016/s1047-3203(03)00019-1 article EN Journal of Visual Communication and Image Representation 2003-05-13
Coming Soon ...