Nachwa Bakr

Nachwa Bakr


I am a Postdoc researcher at ANIMA Team, since Oct 2020, passionate about machine learning, and interested in applying deep learning techniques to various problems in computer vision and computer graphics.





If you are looking for an internship do not hesitate to contact me and discuss possible opportunities.


2020 – 2021: Postdoc researcher – Inria Grenoble Alpes.

During my postdoc, I am interested in Animating 3D antique statues and engravings.

2016 – 2020: PhD in informatics – University Grenoble Alpes.

During my thesis, I worked on the recognition of human manipulation actions. My thesis addresses the problem of recognition, modelling of human manipulation activities. We study modelling manipulation actions as statetransformations. We describe results on three problems: (1) the use of transferlearning for simultaneous visual recognition of objects and object states, (2) the recognition of manipulation actions from state transitions, and (3) the use of reversibleactions as data augmentation technique for manipulation action recognition. [Manuscript][slides]

2015 – 2016: Masters in informatics – University Grenoble Alpes

My master research topic is action recognition from RGB-D videos. The project evaluates features extracted around densely sampled trajectories and others extracted around motion fields which are generated during video compression. In addition, it studies adding trajectory-based HOG of depth videos around the same interest points for the task of action recognition.

2007 – 2012: Engineering degree in Informatics – Aleppo University

Graduation project: Hand gestures to remotely control presentations. The software detects user’s hand using skin color formula and Haar characteristics. Then it tracks hand trajectory to recognize hand gesture by Hidden Markov Models. Recognized gestures are used to remotely handle a PowerPoint presentation and an image viewer application using a webcam.


  • Aboubakr, Nachwa, James L. Crowley, and Remi Ronfard. “Recognizing Manipulation Actions from State-Transformations.” The forth international workshopon Egocentric Perception, Interaction and Computing at EPIC@CVPR19 (2019). [Project Page], [Paper], [Technical report]
  • Nachwa Aboubakr, Rémi Ronfard, James Crowley. Recognition and Localization of Food in Cooking Videos. CEA/MADiMa’18, Jul 2018, Stockholm, Sweden. 2018, 〈10.1145/3230519.3230590〉. Ground-truth annotation of key-frames in 50 salads dataset. [Project: Project page], [paper], [slides]
  • Nachwa Abou Bakr, James Crowley. Histogram of Oriented Depth Gradients for Action Recognition. ORASIS 2017, Jun 2017, Colleville-sur-Mer, France. pp.1-2, 2017, 〈 [paper]