Research Highlight

Robotics | Man-Machine Interface | Motion Detection

Human Motion Prediction using Learning Method for Human-Robot Collaboration In Assembly Tasks

What if the robot could observe human motions analyze them, and predict their motions and intentions? This will enable robot to work in close proximity with humans and opens new possibilities for customized and flexible Assembly and manufacturing processes. Read more…

3D Dimension Extraction from a Scanned Hand for Design and Modeling of Hand Prosthesis using Deep-Learning Methods

A new method for dimensions extraction from three-dimensional scans that allows the creation of personalized hand-prosthesis without additional engineering design, using a DNN for dimension extraction and adjusting relevant dimensions to a 3D CAD model. Read more…


Design and Motion Planning of Multi-Robot Assembly  Cells for Body-in-White Spot Welding

The design of the multi-robot cell for spot welding relies on two main steps: cell design and off-line multi-robot motion planning. The approach aims at defining a methodology for optimizing the cell design while reducing time and error due to the lack of integration between the design and the motion planning.
Read more…
Bio analysis using Learning Methods

Computational diagnosis of Diabetic Macular Edema using deep learning techniques on Optical Coherence Tomography data

Promoting the clinical care of DME, by computational-based quantitative and qualitative analysis of OCT. Applying novel deep-learning methods on OCT data, including volumetric and longitudinal data. Read more…

Microstructure design and analysis of bio-degradable bone scaffolds

Optimal design of bio-degradable bone scaffolds taking into consideration the scaffolds’ mechanical strength over time as it degrades while living bone cells grow upon it. This geometrical change of shape along with the non-trivial structure makes calculation of structural integrity an intricate task. Read more…

Personalized Medicine: 3D Hierarchical Geometric Modeling and Multiscale Mechanical Analysis for Bone tissue

Closing the gap between the classic homogenization approach that is applied to macro-scale models and the modern micro-finite element method that is applied directly to micro-scale high-resolution models. Read more…
3D Parametric Modeling using Learning Methods


Localization of non-rigid objects in a 3D scene for assembly using learning methods

This work present a methodology of a real time pipeline for identification of orientation and location of non-rigid objects in a 3D scene for assembly by using learning methods. The method handles RGB-D images to extract skeletal points using two convolutional neural networks.

Automatic Calculation of Physical Dimensions of Objects from a RGBD Image Using Deep Learning Methods

In this work, we demonstrate a deep learning algorithm capable of automatically extracting the physical dimensions of a mechanical part in a depth image. The algorithm leverages the architecture of CNNs for 2D images, such as ResNet, in order to extract the dimensions.
Read more…

Alignment of 3D scanned objects using learning methods on Point Clouds

This research aims to perform analysis of scanned 3D Objects using learning methods. The analysis process inherently includes acquisition and preparation of the data and construction of suitable learning method. Read more…

Denoising 3D Object Point Clouds using Deep-Learning Techniques

In this work we present a new deep-learning based approach which performs point cloud denoising without requiring additional clean data. Instead, we assume that
the type of the object described in the point cloud is given. Read more…

Extraction of cuboids and bounding box dimensions from scanned objects using Deep Learning methods

In this research, we propose a deep learning-based approach that extracts the dimensions of cuboids and bounding box from scanned objects that are represented by point clouds. Read more…
3D Segmentation and Classification using Learning Methods

Classification, Segmentation, and Geometric Analysis of 3D Point Clouds using Deep Learning

A new generic approach for over-segmentation in 3D point cloud data in general and specifically adjusted for bone porous micro-structures and their unique properties. Read more…

Fractographic Image Analysis Using Computer Vision and Deep Learning Methods

Computerizing one of the most challenging and time-consuming stages of the fractographic analysis process: locating fatigue striations on SEM (Scanning Electron Microscope) images. Read more…

3D Point Cloud Registration for Localization Using a Deep Neural Network Auto-Encoder

A novel methods for registration between a large-scale point cloud and a close-proximity scanned point cloud, providing a localization solution that is fully independent of prior information. Read more…
Design for Additive Manufacturing

Extrusion-based bio-mimetic shape Hamiltonian tool path for scaffolds design