Jiteng Mu

Johns Hopkins University

M.S. Student

About Me

Currently, I am a second year robotics master student at the Johns Hopkins University(JHU), where I am lucky to work with Dr. Alan Yuille and Dr. Russell H. Taylor. My research lies in the intersection of computer vision, robotics and machine learning.

Before coming to JHU, I received a B.E. in mechanical Engineering in 2017 from Shandong University(SDU), China. From 2014 to 2015, I studied as an exchange student at Huazhong University of Science and Technology(HUST). During the summer of 2016, I interned at the University of Alberta in Canada, sponsored by the China Scholarship Council. I received Chiang Chen Overseas Fellowship(Mainland China) in 2017.

Highlight Projects

project name

Explainable Neural Networks

Compositionality is believed to be fundamental to human vision. Compositional models represent visual patterns as hierarchies of meaningful visual concepts. By decomposing complex patterns into simple pieces, people can easily find the logic behind patterns. In this project, we aim to develop networks with better interpretability. (ongoing)

project name

Development of a Mosquitoes Dissection Robot

The project was motivated to address a bottleneck stage of the production of malaria vaccines using mosquitoes, the salivary gland extraction. Currently, the extraction process is completely done by well-trained operators. In order to increase the productivity and make vaccines more affordable, this project aimed to build a fully automated mosquitoes dissection robot capable of grasping mosquitoes by their proboscises.

project name


The ball recognition algorithm in Robocup is challenging with regard to the computing complexity and false positive rate. In this project, we developed an efficient real-time computer vision algorithm for a soccer robot for the 19th Robocup. Our recognition method combines a contour feature method and modified gradient Hough Circle Transform so it can take advantages from both.

Find out more

Other Projects

project name

Seamless Image Cloning with Semantic Segmentation

Cropping out your profile in a portrait photo and blending it into a new background is tedious and time-consuming when done manually. In this project, we propose a two-step solution to this kind of scenario where we use FCN to automatically extract objects from a source image and seamlessly blend them into the given background by Poisson Image Editing. (Course project)

View on GitHub

project name

Classification with the Siamiese Network

In this project, we trained a Siamese network to output two encodings, then we compare these two encodings to check whether there is a similarity between the two images. Based on the similarity of two encodings, we compare and tell if the two images have the same person or not. We explored different distance functions to see their effects in the network accuracy. (Course project)

project name

Hybrid Plane

This project is done during my internship at the University of Alberta. In this project, we designed and tuned the hybrid plane, a combination of a fixed wing plane and a coaxial copter. The new configuration gives the the aircraft more flexibility and dexterity compared to exsited configurations. The project's outcome was presented in the intern symposium at University of Alberta.


A new efficient real-time arbitrary colored ball recognition method for a humanoid soccer robot

Jiteng Mu, Yunxuan Li

The 12th World Congress on Intelligent Control and Automation (WCICA 2016), Guilin, China