Skip links to main content
Logo of Geometric Image Processing Laboratory,
Computer Science Department Home
Technion Home Page

Projects Proposed Projects

Please see propose projects for Spring and Winter upcoming semesters. This is a 4 academic point course. For further details please contact the laboratory engineer Yaron Honen (04-8295535, room 441).

Project Title:

Deep Semantic Shape Reconstruction

Picture of Deep Semantic Shape Reconstruction
Abstract:

Deep Learning architectures have revolutionized the field of Computer Vision, completely altering the way we view 2D and 3D signals. In this project, we tackle the problem of precise and realistic reconstruction of 3D signals by semantic segmentation. The student will take part in creating novel architectures to push current state-of-the-art benchmarks one step further. More details: @https://arxiv.org/abs/2001.09650

Supervisors:
Requirements:
Some background in Image and Signal Processing. An elementary course in Deep Learning (e.g. 236605) and hands-on experience with Tensorflow/PyTorch/Keras. The project is reserved for dedicated excellent students with exceptional hands-on programming and system engineering skills. It is research-oriented project
Project Title:

Deep Precise Shape Reconstruction

Picture of Deep Precise Shape Reconstruction
Abstract:

Deep Learning architectures have revolutionized the field of Computer Vision, completely altering the way we view 2D and 3D signals. In this project, we tackle the problem of precise and realistic reconstruction of 3D signals via supervised and unsupervised learning methods. The student will take part in creating novel architectures to push current state-of-the-art benchmarks one step further. More details: @https://arxiv.org/abs/2001.09650

Supervisors:
Requirements:
Some background in Image and Signal Processing. Elementary course in Deep Learning (e.g. 236605) and hands-on experience with Tensorflow/PyTorch/Keras. The project is reserved for dedicated excellent students with exceptional hands-on programming and system engineering skills. It is research oriented project.
Project Title:

Virtual Try On

Picture of Virtual Try On
Abstract:

Deep Learning architectures have revolutionized the field of Computer Vision, completely altering the way we view 2D and 3D signals. In this project, we tackle the problem of precise and realistic reconstruction of 3D signals by semantic segmentation. The student will take part in creating novel architectures to push current state-of-the-art benchmarks one step further. More details: @https://arxiv.org/abs/2001.09650

Supervisors:
Requirements:
Some background in Image and Signal Processing. An elementary course in Deep Learning (e.g. 236605) and hands-on experience with Tensorflow/PyTorch/Keras. The project is reserved for dedicated excellent students with exceptional hands-on programming and system engineering skills. It is research-oriented project.
Project Title:

ZoomEmotion

Picture of ZoomEmotion
Abstract:

In this project, we would like to develop the ZoomEmotion - an algorithm that detects the emotional state of the child in real time and signals the teacher when the child is experiencing a negative emotion. In this project we would like to expand the previous project that dealt with facial expreation to recognize the faces of the children and their emotional state in real time. We would also like to quantify the time that negative emotions are present and to provide a feedback

Supervisors:
Requirements:
  • Excellence students
  • Programming experience in C++ and/or Python
  • Advantage courses in Machine/Deep Learning and/or Image Processing
Project Title:

Neuro feedback based Virtual reality for practicing mindfulness principles

Picture of Neuro feedback based Virtual reality for practicing mindfulness principles
Abstract:

In this project we would like to expend the previous project by adding more channels of physiological input from the user together with more sensory feedback to the user. In addition new VR scenes will be designed. 
The project’s main objectives are -  Getting to know the current system. Integrate new physiological inputs of Heart rate and breathing (which are additional outputs of the muse 2) to the system software. integrate a tactile sensory feedback to the user.  Create new VR scenes.

Supervisors:
Requirements:
  • Excellence students
  • Programming experience in C++ and/or Python 
  • Advantage courses in Machine/Deep Learning and/or Image Processing
Project Title:

Learning based mother-child movement synchronization

Picture of Learning based mother-child movement synchronization
Abstract:

Mother-child joint storytelling is known to promote the child's cognitive and emotional skills. Behavioral \ physiological synchrony between parent and child during a joined interaction can give a quantified parameter of their relationship and indicate the interaction session's quality. We can evaluate synchrony between mother and child during mutual interaction sessions using various behavioral and physiological measures. In this project, we will focus on mother-child movement synchronization during collaborative storytelling. We will use an existing data set of videos; each has a recording of a mother telling stories to her child. 

Objectives  1. Recognize mother vs. child. 2. Detect of the whole body area of the mother and the child pairs. 3. Quantify movement patterns of both mother and child as a function of time. 4. Calculate the correlation between the movement patterns of the mother and the movement patterns of the child

Supervisors:
Requirements:
· Excellence students 
· Programming experience: C++ and or Python 
· Deep learning techniques
Project Title:

Analysis of teachers movements in classroom

Picture of Analysis of teachers movements in classroom
Abstract:

Using videos of apprentice teachers in classroom, we will inspect their body movements and face orientation. Quantitative and qualitative results will help them perform better in class.

Supervisors:
Requirements:
One of the following: course in geometry, image processing/computer vision, 3D graphics or excellent students.
Project Title:

Computer Vision Based Bike Fit

Picture of Computer Vision Based Bike Fit
Abstract:

In this project we will implement a system that can find the right parameters for adapting the bicycle to the rider.  We will use 3D or 2D camera to capture the scene and than analyze the angles and distances. 





Supervisors:

Alon Zvirin

Yaron Honen

Requirements:
Bicycle fan, bike  rider  and one of the followings: course in geometry, image processing/computer vision, 3D graphics or excellent students 
Copyright © 2016 by Geometric Image Processing Lab. All rights reserved.