PhD Student of Robot Vison and Localization

Nanyang Technological University


Zhang Handuo is currently a Ph.D student in Nanyang Technological University. His research interest is localization and machine learning on robot vision. Currently he is involved in a stereo vison based project for unmanned ground vehicle.


  • Artificial Intelligence
  • 3D Robot Vision including Object Detection and Segmentation
  • Visual Localization and Navigation


  • PhD in 3D Robot Vision, 2016

    Nanyang Technological University

  • MEng in Pattern Recognition, 2011

    Northeastern University

  • BSc in Automation, 2007

    Northeastern University


C & C++




Machine Learning




Project Officer

Nanyang Technological University

Aug 2015 – Jan 2016 Singapore

Responsibilities include:

  • Camera calibration of large baseline stereo cameras.
  • Implement visual SLAM system on stereo camera robot.

Assistant Researcher

Shenyang Institute of Automation, Chinese Academy of Sciences (SIA)

Jul 2013 – Aug 2015 Shenyang, China

Assistance Research in the State Key Laboratory of Robotics, in charge of:

  • System simulatioin for collision detection of robot arms. In charge of algorithm part.
  • Bird-eye-view collision detection system depoyed on robot to construct a panarama environment perception. In charge of algorithm part.
  • Robot monitoring and controlling console system. In charge of programming using C to handle communication between robot and host machine. .

Recent Posts

Seq2seq Model with Attention

Digested and reproduced from Visualizing A Neural Machine Translation Model by Jay Alammar. Table of Contents Sequence-to-sequence models are deep learning models that have achieved a lot of success in tasks like machine translation, text summarization, and image captioning.

How to remotely edit your project without having to use VIM

Remotely editing your work when your server does not have public IP address and you don’t want to spend any money is not so easy. Maybe you can use Team viewer or Anydesk or even chrome remote desktop, but there are high latencies.

Install new linux environment

When you want to install a brand new Ubuntu 16.04 system. You could try to follow this guidance. Open Software & Updates and choose the fastest source. Update the system:

Price and spec of cloud based GPU

I summarized several cloud based GPU services: Name of services Specification Price (US$) AWS P2 instance p2.xLarge 0.9 / hour Azure NC6 1xK80 0.

You only look once (YOLO) -- (2)

YOLO has higher localization errors and the recall (measure how good to locate all objects) is lower, compared to SSD. YOLOv2 is the second version of the YOLO with the objective of improving the accuracy significantly while making it faster.


Stereo Vision System on UGV

Stereo Vision System on High Speed Unmanned Ground Vehicle.

Recent Publications

Quickly discover relevant content by filtering publications.

Multiple Object Tracking With Attention to Appearance, Structure, Motion and Size

Objective of multiple object tracking (MOT) is to assign a unique track identity for all the objects of interest in a video, across the …

GMC: Grid Based Motion Clustering in Dynamic Environment

Conventional SLAM algorithms takes a strong assumption of scene motionlessness, which limits the application in real environments. This …

Heading Reference-Assisted Pose Estimation for Ground Vehicles

In this paper, heading reference-assisted pose estimation (HRPE) has been proposed to compensate inherent drift of visual odometry (VO) …

Ultra-wideband aided fast localization and mapping system

This paper proposes an ultra-wideband (UWB) aided localization and mapping system that leverages on inertial sensor and depth camera. …

A Hybrid Feature Parametrization for Improving Stereo-SLAM Consistency

In visual simultaneous localization and mapping (SLAM) field, especially for feature based stereo-SLAM, data association is one of the …