Paper

Paper is a good way to communicate with others in some areas.

Google Scholar: Link

Research Field: Recommendation System, Video-related Work (Action Recognition, Video Captioning, Video Attribute Recognition), Metric Learning (old)


Dual Dynamic Threshold Adjustment Strategy
Early Access PDF ACM Transactions on Multimedia Computing, Communications and Applications (ToMM),JCR-1区
X Jiang, Y Yao, S Liu, F Shen, L Nie, XS Hua

发表时间: 2024.4
参与程度: ⭐ ⭐ ⭐ ⭐
核心要点:动态自调整对比学习中的margin超参数

Deep Metric Learning with Meta-Mining Strategy with Semiglobal Information
PDF IEEE Transactions on Neural Networks and Learning Systems (TNNLS),JCR-1区
Xiruo Jiang; Sheng Liu; Xili Dai; Guosheng Hu; Xingguo Huang; Yazhou Yao; Guo-Sen Xie; Ling Shao

发表时间: 2022.9
参与程度: ⭐ ⭐ ⭐ ⭐ ⭐
核心要点:利用元学习方法,调整样本对的权重

Bidirectional Maximum Entropy Training with Word Co-occurrence for Video Captioning
PDF IEEE Transaction on Multimedia (TMM),JCR-1区
Sheng Liu, Annan Li, Jiahao Wang, Yunhong Wang

发表时间: 2022.5
参与程度: ⭐ ⭐ ⭐ ⭐ ⭐

核心要点:在视频字幕生成的强化学习方法中加入双向最大熵和词共现性,提升描述质量和多样性

Few-Shot Fine-Grained Action Recognition via Bidirectional Attention and Contrastive Meta-Learning
PDF ACM Multimedia (MM) 2021,CCF A类学术会议
Jiahao Wang, Yunhong Wang, Sheng Liu, Annan Li

发表时间: 2021.07
参与程度: ⭐ ⭐ ⭐ ⭐
核心要点:利用显著性作为注意力,用TPN提取视频特征,研究视频的细粒度少样本分类问题

MiniMCTAD: Minimalist Monte Carlo Transport Architecture Design
PDF IEEE ISPA 2021, CCF C类学术会议
Siqing Fu, Tiejun Li, Jianmin Zhang, Sheng Ma, Sheng Liu

发表时间: 2021.7
参与程度: ⭐

Comparative Study of Intrusion Detection Methods Based on Machine Learning
PDF 《信息网络安全》2018, CCF T3类学术期刊(中国科技核心期刊、CSCD核心库、中文核心期刊)
Xiang He, Sheng Liu, Jiguo Jiang

发表时间: 2018.5
参与程度: ⭐ ⭐ ⭐ ⭐ ⭐
核心要点:入侵检测场景下,比较随机森林、支持向量机和多层感知机