Songyang Zhang

Songyang Zhang

PhD Students

ShanghaiTech University

Biography

I’m a first-year PhD student at ShanghaiTech PLUS Group, advised by Xuming He. I received my undergraduate degree from Beihang University, working with Mai Xu. I have also spent time at research labs of TuSimple Inc, Megvii and Tencent Youtu.

Interests

  • Few/Low-shot Learning
  • Graph Neural Networks
  • Video Understanding

Education

  • PhD Program in CS

    2017.9 - now

    ShanghaiTech University

  • BSc in EE

    2013.9 - 2017.6

    Beihang Univeristy

News

🚩 We have one paper on speech recognition accepted by InterSpeech 2020.

🚩 We have one paper on few-shot semantic segmentation accepted by ECCV 2020 (paper).

🚩 We have one paper on semantic alignment accepted by ICCV 2019 (Paper) (Code)

🚩 We have one paper on graph neural network for vision accepted by ICML 2019 (Paper) (Code)

Selected Publications

Transformer with Bidirectional Decoder for Speech Recognition

Attention-based models have made tremendous progress on end-to-end automatic speech recognition(ASR) recently. However, the …
Xi Chen, Songyang Zhang, Dandan Song, Peng Ouyang, Shouyi Yin
In InterSpeech, 2020.
Transformer with Bidirectional Decoder for Speech Recognition

Part-aware prototype Network for Few-shot Semantic Segmentation

Few-shot semantic segmentation aims to learn to segment new object classes with only a few annotated examples, which has a wide range …
Yongfei Liu, Xiangyi Zhang, Songyang Zhang, Xuming He
In European Conference of Computer Vision, 2020.
PDF Code
Part-aware prototype Network for Few-shot Semantic Segmentation

Dynamic Context Correspondence Network for Semantic Alignment

We instantiate our strategy by designing an end-to-end learnable deep network, named as Dynamic Context Correspondence Network (DCCNet), for the semantic correspondence problem.
Shuaiyi Huang, Qiuyue Wang, Songyang Zhang, Shipeng Yan, Xuming He
In International Conference on Computer Vision, 2019.
PDF Code Poster
Dynamic Context Correspondence Network for Semantic Alignment

Experience

 
 
 
 
 

Research Intern

Megvii Research

Jul 2020 – Present Beijing, China

Research Topics:

  • Object Detection
  • Semantic Segmentation
 
 
 
 
 

Research Intern

Tencent Youtu

Dec 2019 – Jun 2020 Shenzhen, China

Research Topics:

  • Object Detection
  • Few-shot Learning
 
 
 
 
 

Research Intern

Megvii Research

Jun 2019 – Dec 2019 Beijing, China

Research Topics:

  • Object Detection
  • Semantic Segmentation
 
 
 
 
 

Research Inern

TuSimple Inc

Dec 2016 – Jun 2017 Beijing, China

Research Topics:

  • Video Understanding
  • Semantic Segmentation

Recent Posts

PyTorch Tips(FLOPs计算/参数量/计算图可视化/相关性分析)

PyTorch Tips(FLOPs计算/参数量/计算图可视化/相关性分析)
Mar 11, 2020 2 min read
PyTorch Tips(FLOPs计算/参数量/计算图可视化/相关性分析)

Research Topics

Attention Few-shot Learning Few-shot Segmentation Graph Neural Network Meta Learning Object Detection Reinforcement Learning Saliency Semantic Correspondence Spatial Context Speech Recognition Speech Transformer

Contact

  • 201210 Shanghai Pudong Zhangjiang Campus:393 Middle Huaxia Road
  • SIST Building No.1 and take the stairs to Office 311 on Floor 3