Picture for Taojiannan Yang

Taojiannan Yang

AIM: Adapting Image Models for Efficient Video Action Recognition

Feb 06, 2023
Taojiannan Yang, Yi Zhu, Yusheng Xie, Aston Zhang, Chen Chen, Mu Li

Figure 1 for AIM: Adapting Image Models for Efficient Video Action Recognition

Figure 2 for AIM: Adapting Image Models for Efficient Video Action Recognition

Figure 3 for AIM: Adapting Image Models for Efficient Video Action Recognition

Figure 4 for AIM: Adapting Image Models for Efficient Video Action Recognition

* Accepted to ICLR 2023. Project webpage is at https://adapt-image-models.github.io/

Access Paper or Ask Questions

Revisiting Training-free NAS Metrics: An Efficient Training-based Method

Nov 16, 2022
Taojiannan Yang, Linjie Yang, Xiaojie Jin, Chen Chen

Figure 1 for Revisiting Training-free NAS Metrics: An Efficient Training-based Method

Figure 2 for Revisiting Training-free NAS Metrics: An Efficient Training-based Method

Figure 3 for Revisiting Training-free NAS Metrics: An Efficient Training-based Method

Figure 4 for Revisiting Training-free NAS Metrics: An Efficient Training-based Method

* Accepted to WACV2023

Access Paper or Ask Questions

BDANet: Multiscale Convolutional Neural Network with Cross-directional Attention for Building Damage Assessment from Satellite Images

May 16, 2021
Yu Shen, Sijie Zhu, Taojiannan Yang, Chen Chen, Delu Pan, Jianyu Chen, Liang Xiao, Qian Du

Figure 1 for BDANet: Multiscale Convolutional Neural Network with Cross-directional Attention for Building Damage Assessment from Satellite Images

Figure 2 for BDANet: Multiscale Convolutional Neural Network with Cross-directional Attention for Building Damage Assessment from Satellite Images

Figure 3 for BDANet: Multiscale Convolutional Neural Network with Cross-directional Attention for Building Damage Assessment from Satellite Images

Figure 4 for BDANet: Multiscale Convolutional Neural Network with Cross-directional Attention for Building Damage Assessment from Satellite Images

* arXiv admin note: text overlap with arXiv:2010.14014

Access Paper or Ask Questions

MutualNet: Adaptive ConvNet via Mutual Learning from Different Model Configurations

May 14, 2021
Taojiannan Yang, Sijie Zhu, Matias Mendieta, Pu Wang, Ravikumar Balakrishnan, Minwoo Lee, Tao Han, Mubarak Shah, Chen Chen

Figure 1 for MutualNet: Adaptive ConvNet via Mutual Learning from Different Model Configurations

Figure 2 for MutualNet: Adaptive ConvNet via Mutual Learning from Different Model Configurations

Figure 3 for MutualNet: Adaptive ConvNet via Mutual Learning from Different Model Configurations

Figure 4 for MutualNet: Adaptive ConvNet via Mutual Learning from Different Model Configurations

* Extended version of arXiv:1909.12978. More experiments on 3D networks (SlowFast, X3D) and analyses on training cost

Access Paper or Ask Questions