Picture for Yibo Yang

Yibo Yang

Asymmetrically-powered Neural Image Compression with Shallow Decoders

Apr 13, 2023
Yibo Yang, Stephan Mandt

* Preprint

Access Paper or Ask Questions

Neural Collapse Inspired Federated Learning with Non-iid Data

Mar 31, 2023
Chenxi Huang, Liang Xie, Yibo Yang, Wenxiao Wang, Binbin Lin, Deng Cai

Figure 1 for Neural Collapse Inspired Federated Learning with Non-iid Data

Figure 2 for Neural Collapse Inspired Federated Learning with Non-iid Data

Figure 3 for Neural Collapse Inspired Federated Learning with Non-iid Data

Figure 4 for Neural Collapse Inspired Federated Learning with Non-iid Data

* 11 pages, 5 figures

Access Paper or Ask Questions

General Rotation Invariance Learning for Point Clouds via Weight-Feature Alignment

Feb 20, 2023
Liang Xie, Yibo Yang, Wenxiao Wang, Binbin Lin, Deng Cai, Xiaofei He

Figure 1 for General Rotation Invariance Learning for Point Clouds via Weight-Feature Alignment

Figure 2 for General Rotation Invariance Learning for Point Clouds via Weight-Feature Alignment

Figure 3 for General Rotation Invariance Learning for Point Clouds via Weight-Feature Alignment

Figure 4 for General Rotation Invariance Learning for Point Clouds via Weight-Feature Alignment

* 14 pages, 4 figures

Access Paper or Ask Questions

Neural Collapse Inspired Feature-Classifier Alignment for Few-Shot Class Incremental Learning

Feb 06, 2023
Yibo Yang, Haobo Yuan, Xiangtai Li, Zhouchen Lin, Philip Torr, Dacheng Tao

Figure 1 for Neural Collapse Inspired Feature-Classifier Alignment for Few-Shot Class Incremental Learning

Figure 2 for Neural Collapse Inspired Feature-Classifier Alignment for Few-Shot Class Incremental Learning

Figure 3 for Neural Collapse Inspired Feature-Classifier Alignment for Few-Shot Class Incremental Learning

Figure 4 for Neural Collapse Inspired Feature-Classifier Alignment for Few-Shot Class Incremental Learning

* ICLR 2023 (Notable-top-25%)

Access Paper or Ask Questions

DeMT: Deformable Mixer Transformer for Multi-Task Learning of Dense Prediction

Jan 12, 2023
Yangyang Xu, Yibo Yang, Lefei Zhang

Figure 1 for DeMT: Deformable Mixer Transformer for Multi-Task Learning of Dense Prediction

Figure 2 for DeMT: Deformable Mixer Transformer for Multi-Task Learning of Dense Prediction

Figure 3 for DeMT: Deformable Mixer Transformer for Multi-Task Learning of Dense Prediction

Figure 4 for DeMT: Deformable Mixer Transformer for Multi-Task Learning of Dense Prediction

* Accepted by AAAI2023

Access Paper or Ask Questions

Understanding Imbalanced Semantic Segmentation Through Neural Collapse

Jan 03, 2023
Zhisheng Zhong, Jiequan Cui, Yibo Yang, Xiaoyang Wu, Xiaojuan Qi, Xiangyu Zhang, Jiaya Jia

Figure 1 for Understanding Imbalanced Semantic Segmentation Through Neural Collapse

Figure 2 for Understanding Imbalanced Semantic Segmentation Through Neural Collapse

Figure 3 for Understanding Imbalanced Semantic Segmentation Through Neural Collapse

Figure 4 for Understanding Imbalanced Semantic Segmentation Through Neural Collapse

* Technical Report

Access Paper or Ask Questions

PanopticPartFormer++: A Unified and Decoupled View for Panoptic Part Segmentation

Jan 03, 2023
Xiangtai Li, Shilin Xu, Yibo Yang, Haobo Yuan, Guangliang Cheng, Yunhai Tong, Zhouchen Lin, Dacheng Tao

Figure 1 for PanopticPartFormer++: A Unified and Decoupled View for Panoptic Part Segmentation

Figure 2 for PanopticPartFormer++: A Unified and Decoupled View for Panoptic Part Segmentation

Figure 3 for PanopticPartFormer++: A Unified and Decoupled View for Panoptic Part Segmentation

Figure 4 for PanopticPartFormer++: A Unified and Decoupled View for Panoptic Part Segmentation

* Extension of PanopticPartFormer (ECCV 2022). Code: https://github.com/lxtGH/Panoptic-PartFormer. arXiv admin note: substantial text overlap with arXiv:2204.04655

Access Paper or Ask Questions

Towards Theoretically Inspired Neural Initialization Optimization

Oct 12, 2022
Yibo Yang, Hong Wang, Haobo Yuan, Zhouchen Lin

Figure 1 for Towards Theoretically Inspired Neural Initialization Optimization

Figure 2 for Towards Theoretically Inspired Neural Initialization Optimization

Figure 3 for Towards Theoretically Inspired Neural Initialization Optimization

Figure 4 for Towards Theoretically Inspired Neural Initialization Optimization

* NeurIPS 2022

Access Paper or Ask Questions