Picture for Xinlei Chen

Xinlei Chen

ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders

Jan 02, 2023
Sanghyun Woo, Shoubhik Debnath, Ronghang Hu, Xinlei Chen, Zhuang Liu, In So Kweon, Saining Xie

Figure 1 for ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders

Figure 2 for ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders

Figure 3 for ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders

Figure 4 for ConvNeXt V2: Co-designing and Scaling ConvNets with Masked Autoencoders

* Code and models available at https://github.com/facebookresearch/ConvNeXt-V2

Access Paper or Ask Questions

EurNet: Efficient Multi-Range Relational Modeling of Spatial Multi-Relational Data

Nov 23, 2022
Minghao Xu, Yuanfan Guo, Yi Xu, Jian Tang, Xinlei Chen, Yuandong Tian

Figure 1 for EurNet: Efficient Multi-Range Relational Modeling of Spatial Multi-Relational Data

Figure 2 for EurNet: Efficient Multi-Range Relational Modeling of Spatial Multi-Relational Data

Figure 3 for EurNet: Efficient Multi-Range Relational Modeling of Spatial Multi-Relational Data

Figure 4 for EurNet: Efficient Multi-Range Relational Modeling of Spatial Multi-Relational Data

* Research project paper. arXiv v1: codes and model weights of EurNet for image modeling released

Access Paper or Ask Questions

Exploring Long-Sequence Masked Autoencoders

Oct 13, 2022
Ronghang Hu, Shoubhik Debnath, Saining Xie, Xinlei Chen

Figure 1 for Exploring Long-Sequence Masked Autoencoders

Figure 2 for Exploring Long-Sequence Masked Autoencoders

Figure 3 for Exploring Long-Sequence Masked Autoencoders

Figure 4 for Exploring Long-Sequence Masked Autoencoders

* 11 pages

Access Paper or Ask Questions

On the Importance of Asymmetry for Siamese Representation Learning

Apr 01, 2022
Xiao Wang, Haoqi Fan, Yuandong Tian, Daisuke Kihara, Xinlei Chen

Figure 1 for On the Importance of Asymmetry for Siamese Representation Learning

Figure 2 for On the Importance of Asymmetry for Siamese Representation Learning

Figure 3 for On the Importance of Asymmetry for Siamese Representation Learning

Figure 4 for On the Importance of Asymmetry for Siamese Representation Learning

* 11 pages, CVPR 2022

Access Paper or Ask Questions

Masked Autoencoders Are Scalable Vision Learners

Dec 02, 2021
Kaiming He, Xinlei Chen, Saining Xie, Yanghao Li, Piotr Dollár, Ross Girshick

Figure 1 for Masked Autoencoders Are Scalable Vision Learners

Figure 2 for Masked Autoencoders Are Scalable Vision Learners

Figure 3 for Masked Autoencoders Are Scalable Vision Learners

Figure 4 for Masked Autoencoders Are Scalable Vision Learners

* Tech report. arXiv v2: add more transfer learning results

Access Paper or Ask Questions