About Me

CV      Contact: menglong AT cis.upenn.edu

I'm a PhD student in Computer Information Science, GRASP Lab, University of Pennsylvania, working with Kostas Daniilidis. My research interests are computer vision, robotics and machine learning, specialized in object recognition, human action recognition, visual SLAM and text recognition. I obtained a Bachelor's degree in Computer Science from Fudan University in 2010 and a Master's degree in Robotics from the University of Pennsylvania in 2012.

Publications

Active Deformable Part Models Inference,
M. Zhu, N. Atanasov, G. J. Pappas, and K. Daniilidis,
European Conference on Computer Vision (ECCV), 2014.
[PDF / bibtex / video / project page]
@InProceedings{ZhuADPM2014,
  author    = {M. Zhu and N. Atanasov and G. Pappas and K. Daniilidis},
  title     = {{Active Deformable Part Models Inference}},
  year      = {2014},
  booktitle = {European Conference on Computer Vision (ECCV)}
}         
[Technical Report] Active Deformable Part Models,
M. Zhu, N. Atanasov, G. J. Pappas, and K. Daniilidis,
arXiv:1404.0334 [cs.CV], 2014.
[PDF / bibtex]
@article{ZhuADPM14arxiv,
  author    = {Menglong Zhu and
               Nikolay Atanasov and
               George J. Pappas and
               Kostas Daniilidis},
  title     = {Active Deformable Part Models},
  journal   = {CoRR},
  volume    = {abs/1404.0334},
  year      = {2014},
  ee        = {http://arxiv.org/abs/1404.0334},
  bibsource = {DBLP, http://dblp.uni-trier.de}
}         
Semantic Localization Via the Matrix Permanent,
N. Atanasov, M. Zhu, G. J. Pappas, and K. Daniilidis,
Robotics: Science and Systems (RSS), 2014.
[PDF / bibtex]
@InProceedings{Atanasov_SemanticLocalization_RSS14,
  author = {N. Atanasov and M. Zhu and K. Daniilidis and G. Pappas},
  title = {{Semantic Localization Via the Matrix Permanent}},
  year = {2014},
  booktitle={Robotics: Science and Systems (RSS)}
}         
Single Image 3D Object Detection and Pose Estimation for Grasping,
M. Zhu, K. Derpanis, Y. Yang, S. Brahmbhatt, M. Zhang, C. Phillips, M. Lecce and K. Daniilidis,
International Conference on Robotics and Automation (ICRA), 2014.
[PDF] / bibtex / video / project page]
@article{zhu2014grasping,
  title     = {Single Image 3D Object Detection 
               and Pose Estimation for Grasping},
  author    = {Zhu, Menglong and Derpanis, Konstantinos G 
               and Yang, Yinfei and Brahmbhatt, Samarth 
               and Zhang, Mabel and Phillips, Cody 
               and Lecce, Matthieu and Daniilidis, Kostas}
  booktitle = {International Conference on Robotics and Automation}
  year      = {2014} 
}         
From Actemes to Action: A Strongly-supervised Representation for Detailed Action Understanding,
W. Zhang, M. Zhu and K. Derpanis,
International Conference on Computer Vision (ICCV), 2013.
[PDF / bibtex / video / project page / action dataset]
@inproceedings{zhang2013actemes,
  title     = {From Actemes to Action: A Strongly-supervised 
               Representation for Detailed Action Understanding},
  author    = {Zhang, Weiyu and Zhu, Menglong 
               and Derpanis, Konstantinos G},
  booktitle = {International Conference on Computer Vision},
  pages     = {2248--2255},
  year      = {2013},
}         
Monocular Visual Odometry and Dense 3D Reconstruction for On-Road Vehicles,
M. Zhu, S. Ramalingam, Y. Taguchi and T. Garass,
European Conference on Computer Vision (ECCV), workshop on CVVT , 2012.
[PDF / bibtex / video]
@inproceedings{zhu2012monocular,
  title     = {Monocular visual odometry and dense 3d 
               reconstruction for on-road vehicles},
  author    = {Zhu, Menglong and Ramalingam, Srikumar 
               and Taguchi, Yuichi and Garaas, Tyler},
  booktitle = {European Conference on Computer Vision, 
               Workshops and Demonstrations},
  pages     = {596--606},
  year      = {2012},
}         
Literate PR2: Text detection and recognition for indoor environment,
M. Zhu, K. Derpanis, K. Daniilidis,
Robotics Operating System (ROS), 2011.
[ROS wiki / video]

Patents

Method and System for Determining Poses of Vehicle-Mounted Cameras for In-Road Obstacle Detection, US 20140037136 A1.
M. Zhu, S. Ramalingam and Y. Taguchi, [Link]

Datasets

15-class action dataset, frame-by-frame annotated human body joints.

Code

Fast object recogition system, Active Deformable Part Models.
Video annotation toolbox for human joints, Amazon Mturk deployable.
3+1 Point pose estimation, involved in Google Project Tango.
Open source ROS package for text detection and recognition.
A* planning algorithm with quadtree decomposition of the space in C++.
A light-weight distributed group communication framework in C.

Invited Talks

April 09, 2014, VASC Seminar at CMU