Liang Lin

DDet: Dual-path Dynamic Enhancement Network for Real-World Image Super-Resolution

Feb 25, 2020
Yukai Shi, Haoyu Zhong, Zhijing Yang, Xiaojun Yang, Liang Lin

Figure 2 for DDet: Dual-path Dynamic Enhancement Network for Real-World Image Super-Resolution

Figure 3 for DDet: Dual-path Dynamic Enhancement Network for Real-World Image Super-Resolution

Figure 4 for DDet: Dual-path Dynamic Enhancement Network for Real-World Image Super-Resolution

* Code address: https://github.com/ykshi/DDet

Access Paper or Ask Questions

Depthwise Non-local Module for Fast Salient Object Detection Using a Single Thread

Jan 22, 2020
Haofeng Li, Guanbin Li, Binbin Yang, Guanqi Chen, Liang Lin, Yizhou Yu

Figure 2 for Depthwise Non-local Module for Fast Salient Object Detection Using a Single Thread

Figure 3 for Depthwise Non-local Module for Fast Salient Object Detection Using a Single Thread

Figure 4 for Depthwise Non-local Module for Fast Salient Object Detection Using a Single Thread

* Accepted as a regular paper in the IEEE Transactions on Cybernetics

Access Paper or Ask Questions

Tree-Structured Policy based Progressive Reinforcement Learning for Temporally Language Grounding in Video

Jan 18, 2020
Jie Wu, Guanbin Li, Si Liu, Liang Lin

Figure 2 for Tree-Structured Policy based Progressive Reinforcement Learning for Temporally Language Grounding in Video

Figure 3 for Tree-Structured Policy based Progressive Reinforcement Learning for Temporally Language Grounding in Video

Figure 4 for Tree-Structured Policy based Progressive Reinforcement Learning for Temporally Language Grounding in Video

* To appear in AAAI2020

Access Paper or Ask Questions

Physical-Virtual Collaboration Graph Network for Station-Level Metro Ridership Prediction

Jan 14, 2020
Jingwen Chen, Lingbo Liu, Hefeng Wu, Jiajie Zhen, Guanbin Li, Liang Lin

Figure 2 for Physical-Virtual Collaboration Graph Network for Station-Level Metro Ridership Prediction

Figure 3 for Physical-Virtual Collaboration Graph Network for Station-Level Metro Ridership Prediction

Figure 4 for Physical-Virtual Collaboration Graph Network for Station-Level Metro Ridership Prediction

* Jingwen Chen and Lingbo Liu are co-first authors

Access Paper or Ask Questions

An Adversarial Perturbation Oriented Domain Adaptation Approach for Semantic Segmentation

Dec 18, 2019
Jihan Yang, Ruijia Xu, Ruiyu Li, Xiaojuan Qi, Xiaoyong Shen, Guanbin Li, Liang Lin

Figure 2 for An Adversarial Perturbation Oriented Domain Adaptation Approach for Semantic Segmentation

Figure 3 for An Adversarial Perturbation Oriented Domain Adaptation Approach for Semantic Segmentation

Figure 4 for An Adversarial Perturbation Oriented Domain Adaptation Approach for Semantic Segmentation

* To Appear in AAAI2020

Access Paper or Ask Questions

Blockwisely Supervised Neural Architecture Search with Knowledge Distillation

Nov 29, 2019
Changlin Li, Jiefeng Peng, Liuchun Yuan, Guangrun Wang, Xiaodan Liang, Liang Lin, Xiaojun Chang

Figure 2 for Blockwisely Supervised Neural Architecture Search with Knowledge Distillation

Figure 3 for Blockwisely Supervised Neural Architecture Search with Knowledge Distillation

Figure 4 for Blockwisely Supervised Neural Architecture Search with Knowledge Distillation

* We achieve a state-of-the-art 78.4% top-1 accuracy on ImageNet in a mobile setting, which is about a 2.1% gain over EfficientNet-B0

Access Paper or Ask Questions

Knowledge Graph Transfer Network for Few-Shot Recognition

Nov 21, 2019
Riquan Chen, Tianshui Chen, Xiaolu Hui, Hefeng Wu, Guanbin Li, Liang Lin

Figure 2 for Knowledge Graph Transfer Network for Few-Shot Recognition

Figure 3 for Knowledge Graph Transfer Network for Few-Shot Recognition

Figure 4 for Knowledge Graph Transfer Network for Few-Shot Recognition

* accepted by AAAI 2020 as oral paper

Access Paper or Ask Questions

Generalizing Energy-based Generative ConvNets from Particle Evolution Perspective

Oct 31, 2019
Yang Wu, Pengxu Wei, Xu Cai, Guanbin Li, Liang Lin

Figure 2 for Generalizing Energy-based Generative ConvNets from Particle Evolution Perspective

Figure 3 for Generalizing Energy-based Generative ConvNets from Particle Evolution Perspective

Figure 4 for Generalizing Energy-based Generative ConvNets from Particle Evolution Perspective

* To be appeared in T-PAMI 2020. We track and rederive the origin of the pGCN model from the perspective of particle evolution and discover the potential factors that may lead to a deteriorating quality of generated samples and instability of model training. The source code is publicly available at https://github.com/uiyo/GeneralizedGCN

Access Paper or Ask Questions