Moustafa Meshry
Email: MMESHRY [AT] CS [DOT] UMD [DOT] EDU
I'm a PhD candidate at the Perception and Intelligence Lab at the University of Maryland, College Park. I'm advised by Larry Davis and Abhinav Shrivastava. My research interests lie at the intersection of computer vision and graphics for photo-realistic image synthesis. My research focuses on improving the training and quality of controllable generative models (e.g., multi-modal image-to-image translation), as well as applying neural rendering techniques for the task of novel view synthesis. I'm particularly excited about the combination of generative models and neural implicit representations for photo-realistic and 3D-aware image synthesis.
Latest news
- 2021-11-23: We present RTMV, a large-scale ray-traced synthetic dataset for novel view synthesis.
- 2021-07-22: Our Learned Spatial Representation (LSR) paper has been accepted to ICCV, 2021.
- 2021-06-20: Code for StEP has been released.
- 2021-05-31: I'm excited to join Nvidia Research as a research intern working on real-time neural rendering.
- 2021-04-29: We present LSR, a power latent spatial representation for few-shot talking heads.
- 2021-03-03: Our StEP paper has been accepted to CVPR 2021.
Publications
 |
RTMV: A Ray-Traced Multi-View Synthetic Dataset for Novel View Synthesis
Jonathan Tremblay*,
Moustafa Meshry*,
Alex Evans,
Jan Kautz,
Alexander Keller,
Sameh Khamis,
Charles Loop,
Nathan Morrical,
Thomas Müller,
Koki Nagano,
Towaki Takikawa,
Stan Birchfield |
 |
Learned Spatial Representations for Few-shot Talking-Head Synthesis
Moustafa Meshry, Saksham Suri, Larry S. Davis, Abhinav Shrivastava |
 |
StEP: Style-based Encoder Pre-training for Multi-modal Image Synthesis
Moustafa Meshry, Yixuan Ren, Larry S. Davis, Abhinav Shrivastava |
 |
Neural Rerendering in the Wild
Moustafa Meshry, Dan B Goldman, Sameh Khamis, Hugues Hoppe, Rohit Pandey, Noah Snavely, Ricardo Martin-Brualla |
 |
Two Stream Self-Supervised Learning for Action Recognition
Ahmed Taha, Moustafa Meshry, Xitong Yang, Yi-Ting Chen, Larry S. Davis |
 |
Texture Synthesis with Recurrent Variational Auto-Encoder
Rohan Chandra, Sachin Grover, Kyungjun Lee, Moustafa Meshry, Ahmed Taha |
 |
Linear-time Online Action Detection From 3D Skeletal Data Using Bags of Gesturelets
Moustafa Meshry, Mohamed E. Hussein, Marwan Torki Code | IEEE Xplore | ArXiv |
 |
Multi-Modality Feature Transform: An Interactive Image Segmentation Approach
Moustafa Meshry, Ahmed Taha, Marwan Torki |
 |
A scalable maximum-clique algorithm using Apache Spark
Amr Elmasry, Ayman Khalafallah, Moustafa Meshry |
Patents
 |
Few-shot Synthesis of Talking Heads -- U.S. Patent Application No. 17/310,678 - filed April 2022.
Moustafa Meshry, Sofien Bouaziz, Dan B Goldman, Ke Li, Rohit Pandey, Ricardo Martin-Brualla |
 |
Neural rerendering from 3D models. -- U.S. Patent No. 11,288,857 - March, 2022.
Moustafa Meshry, Dan B Goldman, Sameh Khamis, Hugues Hoppe, Rohit Pandey, Noah Snavely, Ricardo Martin-Brualla |
Honours and Awards
- Best paper award finalist at the Conference on Computer Vision and Pattern Recognition (CVPR), 2019 (top 1% of submissions).
- University of Maryland Graduate School Deans Fellowship, 2016 and 2017
- EducationUSA Opportunity Fund: Support from the America-Mideast Educational and Training Services Inc. is awarded to highly qualified students to study in the US.
- ranked 4th at the ACM Egyptian Collegiate Programming Contest 2011, (ECPC’11).
- ranked 19th at Google Code Jam Africa, 2011.