Marwan A. Torki
Associate Professor
Computer and Systems Engineering Department
Faculty of Engineering
Alexandria University
mtorki @ alexu. edu . eg
  Marwan Torki

I am an Associate Professor at the Computer and Systems Engineering Department, Faculty of Engineering, Alexandria University, Alexandria, Egypt. I obtained my PhD in 2011 from the Department of Computer ScienceRutgers The State University of New Jersey, under the supervision of Prof. Ahmed Elgammal

Dr. Marwan Torki is an Associate Professor in the
Computer and Systems Engineering Department at Alexandria University, Egypt.  He received his B.Sc. and M.Sc. in Computer Science from Alexandria University in 2003 and 2006 respectively and his Ph.D. in Computer Science from Rutgers University NJ, USA in 2011. He is advising several undergraduate/graduate students in different research and development projects focusing mainly on computer vision and machine learning tasks.
Dr. Marwan Torki research wide interests are Computer Vision, Natural Language Processing and Deep learning.  

scholar, scopus, dblp, rg,



Book Chapters and Articles

  1. Hamada Rizk, Marwan Torki, Moustafa Youssef: CellinDeep: Robust and Accurate Cellular-based Indoor Localization via Deep Learning. IEEE Sensors Journal 12/2018; PP(99):1-1., DOI:10.1109/JSEN.2018.2885958
  2. Maha Heikal, Marwan Torki, Nagwa El-Makky: Sentiment Analysis of Arabic Tweets using Deep Learning. Procedia Computer Science 01/2018; 142:114-122., DOI:10.1016/j.procs.2018.10.466
  3. Chan-Su Lee, AhmedElgammal and MarwanTorki "Learning Representations From Multiple Manifolds" Pattern Recognition Journal, Feb. 2016 [pdf][doi]
  4. Ahmed Elgammal and Marwan Torki "Learning Image Manifolds from Local Features" Manifold Learning Theory and Applications, Edited by Yun Fu, CRC Press 2011, Pages 233252 Print ISBN: 978-1-4398-7109-6 [pdf]

Conference Papers

  1. Karim Elawad and Mohamed Ezz Eldin and Marwan Torki : DeepCReg: Improving Cellular-based Outdoor Localization Using CNN-based Regressors. (WCNC 2020, 4/2020)[Code, pdf]
  2. Dina Samak, Sarah Jad, Nour Eldin Nabil, Amr Wassal, Marwan Torki: Spectrometer as an Ubiquitous Sensor for IoT Applications Targeting Food Quality, 6th ACS/IEEE International Conference on Computer Systems and Applications AICCSA 2019 November 3rd to November 7th, 2019 [pdf]
  3. Omar S Mossad and Mustafa ElNainay and Marwan Torki: Deep Convolutional Neural Network with {Multi-Task} Learning Scheme for Modulations Recognition, IWCMC 2019 Advances in 5G and Beyond Symposium (IWCMC 2019 5G Symposium, 6/2019. 
  4. Moustafa Abbas, Moustafa Elhamshary, Hamada Rizk, Marwan Torki, Moustafa Youssef: WiDeep: WiFi-based Accurate and Robust Indoor Localization System using Deep Learning. Percom 2019, Japan; 03/2019
  5. Reham Abobeah, Marwan Torki, Amin Shoukry, Jiro Katto: Bi-Directional Attention Flow for Video Alignment. 14th International Conference on Computer Vision Theory and Applications; 03/2019.
  6. Mai Ibrahim, Marwan Torki, Nagwa El-Makky: Imbalanced Toxic Comments Classification using Data Augmentation and Deep Learning. ICMLA; 12/2018
  7. l Moatasem Madani,, Marwan Torkii : Interactive Image Segmentation Using Multimodal Regularized Kernel Embedding. ICMLA 2018, USA; 12/2018A
  8. bdallah Elsaadawy,, Marwan Torkii, Nagwa El-Makky: A Text Classifier Using Weighted Average Word Embedding. International Japan-Africa Conference on Electronics, Communications, and Computations (JAC-ECC) 2018, Alexandria, Egypt; 12/20188
  9. Ahmed Shokry, Marwan Torki, Moustafa Youssef: DeepLoc: A Ubiquitous Accurate and Low-Overhead Outdoor Cellular Localization System. SIGSPATIAL 2018, USA; 11/2018, DOI:10.1145/3274895.3274909
  10. Mahmoud AbdelLatif, Shady AbdelAziz, Ahmed Elmasry, Mohamed Heeba, Mohamed Samir, Marwan Torki : A Supervised Learning Approach Using The Combination of Semantic and Lexical Features for Arabic Community Question Answering. 15th ACS/IEEE International Conference on Computer Systems and Applications AICCSA 2018, Jordan; 11/2018
  11. Marwan Torki: A Document Descriptor using Covariance of Word Vectors. ACL, Australia; 07/2018
  12. Mai Ibrahim, Marwan Torki, and Mustafa ElNainay, "CNN based indoor localization using RSS time-series", IEEE ISCC 2018 [code and data]
  13. Ameer Albaham, Maram Hasanain, Marwan Torki, Tamer Elsayed QweetFinder: Real-time Finding and Filtering of Question Tweets 2017 39th European Conference on Information Retrieval ECIR 2017.
  14. Rana Malhas, Marwan Torki, Rahma Ali, Evi Yulianti, Tamer Elsayed Real, Live, and Concise: Answering Open-Domain Questions with Word Embedding and Summarization. TREC 2016; 11/2016
  15. Sherin Aly, Lynn Abbott, and Marwan Torki, "A Multi-modal Feature Fusion Framework for Kinect-based Facial Expression Recognition using Dual Kernel Discriminant Analysis (DKDA)" IEEE WACV 2016 [pdf]
  16. Moustafa Meshry, Mohamed E. Hussein and Marwan Torki "Linear-time Online Action Detection From 3D Skeletal Data Using Bags of Gesturelets" IEEE WACV 2016 [pdf][code]
  17. Reem Suwaileh, Maram Hasanain, Marwan Torki, Tamer Elsayed: QU at TREC-2015: Building Real-Time Systems for Tweet Filtering and Question Answering. TREC 2015; 11/2015
  18. Moustafa Meshry, Ahmed Taha and Marwan Torki "Multi-Modality Feature Transform: An Interactive Image Segmentation Approach" BMVC 2015, Swansea city, UK [pdf][code]
  19. Ahmed Taha and Marwan Torki "Seeded Laplacian: An Interactive Image Segmentation Approach usning EigenFunctions" ICIP 2015, Quebec city, Canada [pdf][code]
  20. Amr Sharaf, Marwan Torki, Mohamed Hussein and Motaz Elsaban "Real-time Multi-scale Action Detection From 3D Features" WACV 2015 [pdf] [presentation]
  21. Tarek El-Gaaly, Marwan Torki and Ahmed Elgammal "Spatial-Visual Label Propagation for Local Feature Classification" ICPR 2014 (oral) [pdf][presentation]
  22. Mohamed E. Hussein, Marwan Torki, Mohammad A. Gowayyed and Motaz El-Saban "Human Action Recognition Using a Temporal Hierarchy of Covariance Descriptors on 3D Joint Locations"IJCAI 2013 [pdf][poster][Onset and Offset anntoations for MSRC12 gesture data]
  23. Mohammad A. Gowayyed, Marwan Torki, Mohamed E. Hussein and Motaz El-Saban"Histogram of Oriented Displacements (HOD): Describing Trajectories of Human Joints for Action Recognition"IJCAI 2013 (oral)[pdf][poster][presentation]
  24. Tarek El-Gaaly, Marwan Torki, Ahmed Elgammal and Maneesh Singh "RGBD Object Pose Recognition using Local-Global Multi-Kernel Regression"ICPR 2012 [pdf]
  25. Marwan Torki and Ahmed Elgammal "Regression from Local Features for Viewpoint and Pose Estimation" ICCV 2011[pdf]
  26. Marwan Torki and Ahmed Elgammal, "One-Shot Multi-Set Non-rigid Feature-Spatial Matching" CVPR 2010[pdf]
  27. Marwan Torki and Ahmed Elgammal, "Putting Local Features on A Manifold", CVPR 2010[pdf][code]
  28. Marwan Torki , Ahmed Elgammal and Chan-Su Lee "Learning A Joint Manifold Representation From Multiple Data Sets", ICPR 2010.[pdf]

Workshop Papers

  1. Marwan Torki, Maram Hasanain, Tamer Elsayed "QU-BIGIR at SemEval 2017 Task 3: Using Similarity Features for Arabic Community Question Answering Forums". In Proceedings of the 10th International Workshop on Semantic Evaluation,  Sem-Eval 17.

  2. Rana Malhas, Marwan Torki, and Tamer Elsayed.  2016. QU-IR at SemEval-2016 Task 3:  Learning to rank on Arabic community question  answering forums with word embedding. In Proceedings of the 10th International Workshop on Semantic Evaluation,  Sem-Eval 16, San Diego, CA. 

  3. Reham Mohamed, Maha Ragab, Heba Abdelnasser, Nagwa M El-Makky, Marwan Torki: Al-Bayan: A knowledge-based system for Arabic answer selection. . In Proceedings of the 9th International Workshop on Semantic Evaluation,  Sem-Eval 15. [pdf]

  4. Heba Abdelnasser, Reham Mohamed, Maha Ragab, Alaa Mohamed, Bassant Farouk, Nagwa El-Makky, Marwan Torki: Al-Bayan: An Arabic Question Answering System for the Holy Quran. ANLP; 01/2014 [pdf]

  5. Tarek El-Gaaly, Marwan Torki, Ahmed Elgammal, Maneesh Singh: RGBD Pose Recognition using Local-Global Multi-kernel Regression. 2nd GNY Area Multimedia and Vision Meeting; 06/2012 [pdf]

  6. Tarek El-Gaaly, Marwan Torki, Ahmed Elgammal, Maneesh Singh: Multi-Modal RGBD Sensors for Object Grasping and Manipulation. IEEE/RSJ International Conference on Intelligent Robots and Systems, Workshop on; 10/2012  [pdf]


  1. Marwan Torki, "Learning the Manifolds of Local Features and Their Spatial Arrangments", Ph.D Thesis, Rutgers University, NJ, USA summer 2011.[pdf][presentation]
  2. Marwan Torki, "Towards Efficient View Maintenance In Data Warehouses", M.Sc. thesis, Faculty of Engineering, University of Alexandria. August 2006.