Jun Sun
Peking University
64 Papers
347 Citations
Jun Sun is an academic researcher from Peking University. The author has contributed to research in topics: Scalable Video Coding & Video quality. The author has an hindex of 12, co-authored 62 publications. Previous affiliations of Jun Sun include Chinese Academy of Sciences.
Chat about Author
Papers
Novel Statistical Modeling, Analysis and Implementation of Rate-Distortion Estimation for H.264/AVC Coders
TL;DR: Efficient algorithms for the estimation of block-level rate and distortion and a simple and efficient algorithm is proposed for the distortion estimation, which can be efficiently implemented in a low-complexity way.
69
Enhancing Knowledge Tracing via Adversarial Training
Xiaopeng Guo,Zhijie Huang,Jie Gao,Mingyu Shang,Maojing Shu,Jun Sun +5 more
- 17 Oct 2021
TL;DR: Xiaopeng et al. as mentioned in this paper proposed an adversarial training-based knowledge tracing method (ATKT) to enhance the generalization of the original KT model by adding adversarial perturbations to the original interaction embeddings as adversarial examples.
57
One-for-all: An Efficient Variable Convolution Neural Network for In-loop Filter of VVC
TL;DR: An efficient network architecture VCNN is designed that can not only effectively reduce compression artifacts, but also can be adaptive to various QPs and FTs and integrates into VVC as an additional tool of in-loop filters after the deblocking filter.
38
•Proceedings Article
Efficient SIMD optimization of HEVC encoder over X86 processors
Keji Chen,Yizhou Duan,Leju Yan,Jun Sun,Zongming Guo +4 more
- 01 Dec 2012
TL;DR: This paper focuses on the fast implementation of the HEVC encoder over modern Intel x86 processors, and identifies the most time-consuming modules of HM 6.2 encoder, represented by motion compensation, Hadamard transform, sum of difference (SAD/SSD) calculation and integer transform.
Statistical model, analysis and approximation of rate-distortion function in MPEG-4 FGS videos
TL;DR: The generalized Gaussian distribution is employed first to model the DCT coefficients of image data from MPEG-4 fine-granularity scalability (FGS) frame and a rate-distortion (R-D) model is proposed to approximate the actual distortion-rate function.
34