2016
3D ROOM Reconstruction
Description:
Most of the tablets today have touch screen and camera. Using those two we wish to reconstruct a virtual room from a single video taken in the room.
 
Virtually Playing Catch
Description:
A great difficulty of interacting in a virtual reality environment is that the physical actions are usually not as natural as we are used to in the real world. In this project we will attempt challenge this problem and create a virtual environment where two users can throw virtual objects one to the other. The system will consist of two pairs of virtual reality glasses and an external Kinect camera. Using the camera which can analyze the motions of the players, we will send the information to the virtual world and translate it into the physical motions controlling the propelling of objects. The goal will be that the action is as smooth and natural as it is in the real world.
Shelfie
Description:
Shelfie is the new buzz-word in the 3D Printing community - A 3D-printed model of yourself!

In this project we will build a system to scan a human head using a Kinect camera. At the end of the project, beside of being a computer graphics professional, you will be able to print your self in the Labs 3D Printer.

 
Pool side keeper
Description:
Pools are dangerous, we will develop a camera based alarm system to keep your child safe.
An autonomous guide to Israeli plants
Description:
In this project you will develop and implement an autonomous guide to the indigenous Israeli plants. The goal is to classify an untagged image of a flower taken from a camera such as a smartphone camera into one of predefined species.
 
BlackJack cards counting
Description:
Design a system that gets a video stream as input and: Detect new cards when added to the table, Identify the card and Determine the best move.
 
2015
Facial Keypoints Detection
Description:
Facial keypoints detection is an important and central building block in application such as: Tracking faces in images and videos, analysis of facial expression and emotions, medical analysis of diseases, biometric signature and face recognition.
 
4D Point Cloud Visualization GUI
Description:
Point clouds are the basic measurements of 3D systems. We are building a system that captures 3D point clouds over time. In this project we would like to design a user friendly system to visualize this data. Creativity in suggesting new visualization techniques is encouraged. Knowledge in computer graphics and design\artistic capabilities are an advantage.
Manipulating Virtual Objects
Description:
The future of virtual reality is natural interaction. The user wants the ability to control what he sees in a smooth way using hand position and gestures. Using Intels depth camera we will attempt to create a panel in a virtual reality environment which can be interacted with using precise hand motions. We will attempt to test the limit of the precision necessary to operate the panel such as moving it around the virtual space, enlarging and reducing the size and pressing buttons on the panel.
 
Complete Missing Pieces in a Puzzle
Description:
What could be more frustrating from completing a puzzle and finding out that 2 -3 pieces are missing ?

With today\s technology, we can complete the missing piece even without prior knowledge of how the complete image looks like. We can even print the missing pieces in a 3D printer.

Learn facial expressions from an image
Description:
In this project you will try and design an intelligent algorithm for the task of distinguishing between 7 expressions (Angry, Disgust, Fear, Happy, Sad, Surprise, Neutral ). The training Data set is ~35000 face images and the test set is ~5000 image.
 
Distinguish Dogs from Cats
Description:
In this project, you'll write an algorithm to classify whether images contain either a dog or a cat. This is easy for humans, dogs, and cats. Your computer will find it a bit more difficult.
Manipulating 3D Objects Using a 3D Monitor
Description:
3D vision is an exponentially growing market, with a forecast that in the near future 3D gaming will move from being a novelty to a worldwide market. Another field which has caused great hype is Natural User Interface where humans control computers using hand gestures as commands.
 
Forward Motion Deblurring
Description:
In this project we handle a special type of motion blur considering that cameras move primarily forward or backward. Solving this type of blur is of unique practical importance since nearly all car, traffic and bike-mounted cameras follow out-ofplane translational motion.
Virtual Reality Objects Interaction Project
Description:
Getting back to the original Iron Man movie and its awesome 3D holographic experience combined with one of the most famous gestural interfaces in sci-fi, in our 2014 we decided to show that what was just science-fiction 6 years ago is now becoming science.
 
Real Life Temple Run
Description:
Some games may be easy to play since they only require moving fingers across the screen or on a keyboard. The game becomes more difficult, but more interactive and exciting, when the player himself needs to perform the actions. The goal of the project is to create a life-like version of the well known game Temple run. The game will be made physically interactive using a Kinect camera filming the player, and using skeleton extraction the player will command the game using actual jumping/ducking/turning and running simulated by foot motion.
Image Retrieval using Textual Cues
Description:
We will implement an algorithm for the text-to-image retrieval problem based on textual content present in images. Given the recent developments in understanding text in images, an appealing approach to address this problem is to localize and recognize the text, and then query the database, as in a text retrieval problem.
 
Virtual Odyssey
Description:
Virtual Odyssey is a story driven video game for the PC.
In the game, the player controls a young student who finds himself in a dystopic future, in the body of a robot. It involves various minigames such as navigating a maze, blocking incoming projectiles, and more ? while the mystery behind the enigmatic events of the future unfolds.
Immersive VR 3D minesweeper
Description:
Re-thinking the classic game for the 3D world by creating a fully immersive virtual reality experience combining the Oculus Rift and the Kinect to achieve 6 degrees of freedom and a natural walking sensation.
 
 
2014
Automatic Upright Adjustment of Photographs
Description:
Everyone takes photos while travelling. However, in man-made environment the scene is usually viewed from below and thus distorted. In this project we will build an automatic system for returning the buildings to look normal.
 
3D Cast Modeller
Description:
In this project we will take several photos of a hand from different angles, analyse the received data, design and print a 3D model of a custom cast that will allow making use of the ultrasound technology for accelerating fracture healing. The new cast will also be lighter, waterproof and airy.
Pixelated image abstraction
Description:
What makes pixel art both compelling and difficult is the limitations imposed on the medium. With a significantly limited palette and resolution to work with, the task of creating pixel art becomes one of carefully choosing the set of colors and placing each pixel such that the final image best depicts the original subject. This task is particularly difficult as pixel art is typically viewed at a distance where the pixel grid is clearly visible, which has been shown to contribute to the perception of the image.
 
Objects Recognition Using Sparse Representation
Description:
Recognizing an object is a basic action human beings are doing daily with very small effort. Nevertheless, for a machine, recognition is a complex matter which is being studied thoroughly nowadays.
Key Duplicator
Description:
Have you ever lost your keys? Why not take a photo of your keys and print a copy using a 3D Printer? In this project we will develop an Android application that creates a 3D printer ready copy of a regular key.
 
Multi-user virtual war-room
Description:
Using the newest and most advantaged virtual reality goggles and hand trackers this project will involve creating a virtual war room for use in IDF training demonstrations. The project will involve learning how to use the oculus rift and then building a virtual war room where each soldier or officer sits in a chair in the room and can see the other officers' avatars. Each user has a Razer Hydra and can control various virtual Screens and menus and can communicate using either a keyboard or the microphone.
State of the art image demosiacing
Description:
CMOS sensored digital cameras need to produce an RGB image. However, they only sense one color channel per pixel. Either red, green or blue. To obtain the full RGB image one needs to somehow interpolate over the missing information. This process is known as demosaicing. In this project we shall review and develop advanced demosaicing techniques that have the potential to be efficiently implemented on designated hardware.
 
Aerial image prediction
Description:
Aerial imagery is of a great interest on both civil and military applications. In this project we will explore the possibility of exact prediction of a new image given several previous images. Together with the images the previous and the future locations of the camera are given as well.

This project will be done in cooperation with Elisra.
Nonchronological Video Synopsis and Indexing
Description:
The amount of captured video is growing with the increased numbers of video cameras, especially the increase of millions of surveillance cameras that operate 24 hours/day. Since video browsing and retrieval is time consuming, most captured video is never watched or examined.
 
Meshing natural outdoors scenes from sparse-and-noisy point clouds
Description:
A point cloud is an unorganized set of 3D points, extracted from a scene by an active/passive scan. The meshing challenge is known as reconstructing the original model by covering the points in the point cloud with polygons (triangles, called faces) optimally.
Better sport with Kinect
Description:
Sport is fun. However, most of the time one does not get any feedback about correctness of the performed motions. Would you like the computer to be your personal trainer and tell what is right and what is not? We will use Kinects ability to track human pose to provide such feedback.
 
 
2013
Air Gun Target Scoring
Description:
Air Gun target are traditionally scored manually. A manual process by nature is open for personal interpretation. In this project we will develop an application for Android enabled devices that will scan a target with up to 4 bullets hole and score it. The application will keep track on personal scores of shooters and maintain league results.
 
Video Stabilization for Tablet
Description:
Everyone can shoot video today. This is especially relevant using smartphone or tablet, since we always have one of them. However, amateur video, which one can easily share with friends, usually lacks stability. The reason is that these videos are usually taken from handheld device. In this project we wish to solve this problem by implementing an automatic video stabilization algorithm that can run on a mobile device.
Detection of Unusual Events in Videos
Description:
Real-time unusual event detection in video stream has been a difficult challenge due to the lack of sufficient training information, volatility of the definitions for both normality and abnormality, time constraints, and statistical limitation of the fitness of any parametric models.
 
Virtual Drums
Description:
In this project we will device a NUI system for playing drums. based on skeleton inputs from the kinect we will design a system that displays a virtual drum set (either on a screen or using a projector) and lets a person interact with the drum set by a skeleton-projection interaction.
Eye gaze tracking
Description:
Knowing where the user is looking can be used for various applications, such as eyes interaction, gaze oriented compression, usability tests and many more. Most existing accurate gaze trackers either use cumbersome equipment, expensive hardware or limit the user to keep the head still.
 
Hole In the Wall Game
Description:
The idea is to create a game similar to the crazy Japanese gameshow called Brain Wall.
Sketch a map
Description:
Hand-designed destination maps include only the most important roads in the region and are non-uniformly scaled to ensure that all of the important roads from the highways to the residential streets are visible..
 
Photo Based Travel Guide
Description:
Have you ever wondered what is there to see arouf the corner ? Is there a must see atraction next street ? In this project we will develop an Android application that will display all the photos of near by atractions based on the tablet current Location.
3D Video Chat - Head Tracking
Description:
Video chat and video conference hasn't changed much in the past decade, other than some improvements to speed and video quality. Some systems were created that allowed one user to control the camera direction of the other user, but such systems require cumbersome cameras with motors and are not very intuitive. A new innovative webcam that captures depth data along with image data can be used to create much more natural 3D video chats.
 
Alignment and Mosaicing of Non-Overlapping Images
Description:
Image alignment and mosaicing are usually performed on a set of overlapping images. Without image overlap current methods are helpless, and this is the case we address in this project. So if a traveler wants to create a panoramic mosaic of a scene from pictures he has taken, but realizes back home that his pictures do not overlap, there is still hope.
Android Augmented Reality
Description:
This project is about implementing augmented reality in an Android phone. first phase is creating a software in which when you point the camera on a specific QR code, the software will place a 3d character directly on the code, and it will respond to movements of the phone and camera as if the object was there.
 
Automatic Scene Painting
Description:
What can you do with million images? In this project a new image painting algorithm powered by a huge database of photographs gathered from the Web will be implemented. The algorithm paints a gray scale image by finding similar images in the database.
Look At Me
Description:
Humanoid robots rapidly enter our life. However to communicate efficiently with people a robot need not only to look like a human but also behave accordingly. In this project we will make a robot head look at the face of the person it is interacting with.
 
Comics from Video
Description:
In the project we will build a system for automatic creation of a comics strip from a given video sequence.
Realtime Segmentation using a Tablet
Description:
While the user is taking a video of his surrounding, and the image is displayed on the tablet screen, he can draw a circle inside an object. The algorithm will segment this object out of the frame. The main challenge is performing the segmentation in real time.

This project will be implemented using Lenovo X220 Tablet
 
Richness-Preserving Manga Screening
Description:
Japanese comic, or manga, is a popular art form over the world. Screening refers to the process of laying pre-printed patterns (can be regular or irregular) over a region. \ we will implement a novel method for generating bitonal manga backgrounds from color photographs
Self-Animating Images
Description:
Illusory motion in a still image is a fascinating research topic. Physiologists and psychologists have attempted to understand this phenomenon by constructing simple, \ color repeated asymmetric patterns and have found several useful rules to enhance the strength of illusory motion.
 
Become a professional photographer
Description:
Most of us are not professional photographers. Yet, we would like to take pictures that are equally good as those of professional photographers. In this project we will estimate the quality of a given image in order to give the user several basic instructions. The goal of these instructions is to allow the user to get a better image.
Head tracking 3D viewer
Description:
Nowadays computer animation mostly did in 3D. However, we are still looking at 2D image on TV or computer screen. Even using a 3D capable monitor does not solve this problem completely, since the view point should change when the viewer moves.
 
Moving / static classification via tracking
Description:
Aerial imagery is of a great interest on both civil and military applications. In this project are given a set of aerial images along with many tracks of the objects. The goal is to distinguish with maximal confidence between the tracks that belong to static (background) and to moving objects.

This project will be done in cooperation with Elisra.
3D Model Tracking in a Video
Description:
In this project we will match a 3D model of an object to a video that contains such an object.
 
Video Textures
Description:
A picture is worth a thousand words. And yet, there are many phenomena, both natural and man-made, that are not adequately captured by a single static photo. A waterfall, a flickering flame, a flag flapping in the breeze each of these phenomena has an inherently dynamic quality that a single image simply cannot portray. This project is based on an article written by Arno Schodl et al. that introduces a new type of medium, called a video texture, which has qualities somewhere between those of photograph and a video.
Automatic Photo Album
Description:
Every trip or family event ends with a dilemma. Which pictures out of the hundreds we took should we put in the album? In this project we would like to build an algorithm that decides it automatically.
 
Living in virtual world
Description:
It would be great to live in any world we wish, real or imaginary. In this project we will change the real world (on TV screen however) with virtually any world. The world can be anything from gaming environment to real-size 3D CAD modeling.
Distance Measure using a Tablet
Description:
While the user is taking a video of his surrounding, and the image is displayed on the tablet screen, he can acquire the distance between the tablet to an object by taping on it.

This project will be implemented using Lenovo X220 Tablet
 
Travel Path Planing
Description:
Travling in the open fields is a common way to enjoy The Weekend. In this project we will develop an application that will help the traveler to plan his route from point A to pint B by using topographic infomartion from Gogle map. Thr route should minimze the travler effort.

This project will be done in cooperation with VISL.
Virtual room reconstruction
Description:
Most of the tablets today have touch screen and camera. Using those two we wish to reconstruct a virtual room from a single video taken in the room.
 
 
2012
Gizmo
Description:
Create a 3D model from a live plant.
 
Collision recognition from a video
Description:
Recognize possible collision trajectories by vehicles, using a video taken from a camera directed toward the rear of the direction of driving.
Collision recognition from a video using Android Tablet
Description:
Recognize possible collision trajectories by vehicles, using a video taken from a camera directed toward the rear of the direction of driving.
 
Learning sign language with Kinect
Description:
Sign language is the major communication method for deaf people. However it is not simple for us to learn all the signs. In this project we will build a system for learning American Sign Language (ASL). The system will guide the user and show him video tutorials. Then the user will be requested to reproduce the learned signs.
Implementation of EEG electrodes positions measurement system
Description:
During this project students will implement EEG electrodes positions measurement system. This system implementation will involve: (1) Design of hardware components. (2) Image processing. (3) Implementation of user friendly software.

This project will be done in cooperation with Network Biology Research Lab.
 
Implementation of EEG electrodes positions measurement system
Description:
During this project students will implement EEG electrodes positions measurement system. This system implementation will involve: (1) Design of hardware components. (2) Image processing. (3) Implementation of user friendly software.

This project will be done in cooperation with Network Biology Research Lab.
Registering images to 3D Models
Description:
The project goal is to find an accurate registration between a given 3D model of object and its image. The image is taken by arbitrary camera and in arbitrary lighting conditions. There are several methods for solving this problem. We will implement a system that registering a given image to the 3D model using those methods and test it on a real data.

This project will be done in cooperation with Elbit.
 
Camera path Reconstruction
Description:
Visual navigation is an important part of any automatic vehicle. Furthermore, the using the visual information can increase the accuracy in the position and orientation estimation.

This project will be done in cooperation with Elisra.
Monocular Reconstruction of Inextensible Surfaces
Description:
The ability to reconstruct the 3D structure of a deformable surface from a single viewpoint can lead to a whole range of application in several important fields. In this project we attempt to perform this reconstruction while dealing with its inherit ambiguity by using a local linear deformation model. We also present some basic application for this system.
 
CGMFrame - Computer Graphics/Vision Framework
Description:
In this project we will develop and build a framework to work on Images, Videos and 3D Models - CGMFrame.
The project will be focused on building a user interface and a framework that will enable the view and manipulate (using given, already implemented algorithms) of graphical objects like images, videos and 3D model. Future projects and research papers of the Computer Graphics laboratory will use the CGMFrame.
Automatic Photo Album
Description:
Every trip or family event ends with a dilemma. Which pictures out of the hundreds we took should we put in the album? In this project we would like to build an algorithm that decides it automatically.
 
Simulating Decorative Mosaics
Description:
Artists often invent techniques later used in computer graphics. Tile mosaics, for example, are images made by cementing together small polygonal colored patches. They are early examples of image synthesis techniques such as point sampling and rasters of pixels.
Invertible motion Blur in Video
Description:
The project deals with motion deblurring in successive video frames. The blur can be invertible even if the point-spread function (PSF) due to motion smear in a single photo is non-invertible. The key idea is to record the same object with varying PSFs, so that the nulls in the frequency component of one frame can be filled by other frames.
 
Registration of images from moving camera
Description:
Filming the world from a vehicle can yield a large amount of useful information. However before processing the acquired images it is highly desired to bring them to the same coordinate system. This process usually referred as registration.

This project will be done in cooperation with Elisra.
 
2011
Motion Deblurring in images
Description:
One of the most common artifacts in digital photography is motion blur caused by camera shake. In many situations there simply is not enough light to avoid using a long shutter speed and the inevitable result is that many of our snapshots come out blurry and disappointing.
 
Automatic Joiners
Description:
There are two kinds of Joiners. Joiners of the first kind are created by taking several pictures of a scene from different points of view, and putting them one over the other to form a joint image. Joiners of the second kind are made of squares, taken from different images, and placed one aside the other to form a full image.
Paper Architecture
Description:
Paper architectures are 3D paper buildings created by folding and cutting. The creation process of paper architecture is often labor-intensive and highly skill-demanding, even with the aid of existing computer-aided design tools.
 
Efficient Gaze Correction
Description:
Today webcams are very popular. They can be found in almost every modern laptop. Still there is a common problem in video-chat: the user is usually looking at the center of the display and not to the camera.
Floor plan from panorama creation
Description:
A 360 panorama is a convenient way to convey rooms, houses and other indoor environments. The problem with it that the viewer get confused because he looks in different directions simultaneously
 
Augmented Panoramic Video
Description:
Many video sequences consist of a locally dynamic background containing moving foreground subjects. In this project we will remove the foreground subjects from the original video, the remaining elements will be merged into a dynamic background panorama, which seamlessly extends the original video footage.
Augmented Reality
Description:
The input of the project is a video sweeping a room (only a part of it) and the goal is to add some virtual objects (like furniture) to the frames of the video. We will use structure-from-motion methods to create a simple 3D reconstruction of the room. Then we add the objects and return them back to video (as described in the attached paper). The whole work does not intended to be real-time - we will use Matlab implementation.

This project will be done in cooperation with Elbit.
 
Lego figures from 3D models
Description:
Who doesnt love action figures? In this project we will scan toys in a special 3D scanner and develop an algorithm to generate instructions of how to build the figures from Lego bricks.
From depth camera to 3D video
Description:
3D video has gained an increased interest in the last years. Usually the 3D video produced by filming the scene by two connected cameras. But there is another way to create a 3D video use depth information of the scene.

This project will be done in cooperation with Elbit.
 
Stereo and Time-of-Flight 3D Fusion
Description:
2 of the main method to capture depth info are stereo triangulation and filming using ToF (Time-of Flight) camera.
Each one of the above method has its pros and cons. In this project we will use both of them to get the good of both methods

This project will be done in cooperation with General Motors.
Soccer 3D Reconstruction
Description:
In this project we will reconstruct the 3D positions of the players and the ball from a single video. This algorithm can be later used for redisplaying the game from other angle, for discovering judge mistakes, 3D animation and others.
 
Multi-view reconstruction of dynamic scenes
Description:
In this project we intend to explore the ability of an automatic system to reconstruct 3D human movement using several video streams.
Parallelization of Ray Tracing Units
Description:
This project studies a split multiple RTP (Ray Tracing Processors) ray-tracing, by simulating a system of CPU and dual RTP by three separate PCs connected by LAN.

This project will be done in cooperation with Lucid.
 
Creating coloring page from photo
Description:
Most of the kids love coloring pages with their favorite characters. By why looking for such pages on the web if one can create them automatically from any digital image. Such images can be found easily or even takes from movies.
Cartoon Yourself
Description:
Creating your own cartoon is neat. You can use it as your own avatar on Facebook, Myspace, and other social networks.
 
Automatic Scene Painting
Description:
What can you do with million images? In this project a new image painting algorithm powered by a huge database of photographs gathered from the Web will be implemented. The algorithm paints a gray scale image by finding similar images in the database.
From Image to Painting
Description:
Photos immortalize memories, happy moments and significant event in someones life. Painting can also highlight the emotions of this photo to unfamiliar persons. We wish to implement an aoristic method that automatically renders images to painting.
 
Photo Enhancement
Description:
Combining regions of multiple photographs or videos into a seamless composite is a fundamental problem in many vision and graphics applications, such as image compositing, mosaicing, scene completion, and texture synthesis.
Detecting Texts in Images
Description:
Detecting text in natural images, as opposed to scans of printed pages, faxes and business cards, is an important step for a number of Computer Vision applications, such as computerized aid for visually impaired, automatic geocoding of businesses, and robotic navigation in urban environments. We will develop an algorithm, robust to fonts and language, which detects text in natural images.
 
Advances segmentation of rigid objects
Description:
In this project we will perform a segmentation of a rigid object in a video sequence. We will tell the object from its surroundings using its distinct kinematics.
3D from multiple cameras
Description:
Depth extraction from stereo-pair is a well known challenge. The use of more than 2 cameras however, has a potential to improve system performance in several aspects: Occlusions avoidance, failure reduction, accuracy increase, reflectance map prediction for multiple views.

This project will be done in cooperation with Intel.
 
From sketch to Image
Description:
Finding an image in a database that is close to a mental model is an important and difficult task. Currently, most queries are either based on textual annotations, rough color sketches or other images.
Flower Recognizer
Description:
The ability to automatically recognize the type of a flower from a single image can be very attractive for a casual tourist, especially if it can be performed in real time.
 
 
2010
Shadow Art
Description:
Shadow art is a unique form of sculptural art where the 2D shadows cast by 3D sculpture are essential for the artistic effect.
 
Image Based Material Editing
Description:
Suppose you have a photograph of some object say a bronze sculpture or a porcelain vase and you want to know how it would look if it were made out of some completely different material, such as glass, plastic, or wax.
Interactive Normal Reconstruction from a Single Image
Description:
Our goal is to reconstruct a 3D surface from a single image. This operation is important for modeling and processing software.
 
3D-Modeling by Ortho-Image Generation from Image Sequences
Description:
Modeling from images is a very important and popular tool. We will implement a semi-automatic approach that enables the generation of a high-quality 3D model of a static object from an image sequence that was taken by a moving, uncalibrated consumer camera.
Stipple drawing
Description:
Creating drawings by using mostly dots is a powerful and widely used illustration method. Typically, several tens of thousands of dots are manually arranged to generate a single drawing...
 
Automating Joiners
Description:
Joiners is a term coined by the artist David Hockney. Joiners are made of squares, taken from different images, and placed one aside the other to form a full image...
Intrinsic Colorization
Description:
In this project, we present an example-based colorization technique robust to illumination differences between grayscale target and color reference images.
 
Symmetry detection
Description:
3D Euler spirals are visually pleasing, due to their property of having their curvature and their torsion change linearly with arc-length.
Shift-Map Image Complition
Description:
Geometric rearrangement of images includes operations such as image retargeting, inpainting, or object rearrangement. Each such operation can be characterized by a shiftmap: the relative shift of every pixel in the output image from its source in an input image.
 
Image Super-Resolution using Gradient Profile Prior
Description:
We will implement a super-resolution approach using a novel generic image prior gradient profile prior, which is a parametric prior describing the shape and the sharpness of the image gradients.
Detail Preserving Shape Deformation in Image Editing
Description:
Shape deformation is a common practice in digital image editing, but can unrealistically stretch or compress texture detail. We propose an image editing system that decouples feature position from pixel color generation, by resynthesizing texture from the source image to preserve its detail and orientation around a new feature curve location.
 
Super-Resolution from a Single Image
Description:
The goal of Super-Resolution methods is to recover a high resolution image from one or more low resolution input images. Our approach is based on the observation that patches in a natural image tend to recur many times inside the image within and across scales.
Edge-Preserving Decompositions
Description:
The problem of image interpolation (producing a sequence of plausible intermediate frames between two input images) has a wide range of applications like animation of still images, temporal interpolation for up-sampling of lowframe rate videos, and interpolating viewpoints between imagescaptured with different camera locations
 
Scene painting
Description:
Coloring grayscale images is a challenging task. Given no information on the color, any grayscale pixel can represent several possible colors. In this project we will develop a method for coloring a grayscale image in a spatial consistent manner.
Image Completion with Structure Propagation
Description:
Completing unknown "holes" in an image is a challenging task. In this project a novel approach to image completion will be tested. In this approach user-specified curves are used to form structure propagation. The remaining unknown regions are filled using patch-based texture synthesis.
 
Removing Image Artifacts
Description:
Dirt on camera lenses, and occlusions from thin objects such as fences, are two important types of artifacts in digital imaging systems. These artifacts are not only an annoyance for photographers, but also a hold-up to computer vision.
Single Image Dehazing
Best lab Project of 2010
Description:
The ability to obtain lighting information from a single image seems unrealistic, but is really possible.We will implement a new method for estimating the optical transmission in hazy scenes given a single input image. Based on this estimation, the scattered light is eliminated to increase scene visibility and recover haze-free scene contrasts.
 
3D Measurement using structured light
Description:
We would like to be able to capture a full color video and at the same rate to obtain a depth map of the scene captured.

This project will be done in cooperation with Intel.
 
2009
Virtual Mouse
Description:
In this project, we will develop a way to find this transformation using 3D camera. We will create a virtual mouse to exemplify the transformation.
 
Second Life
Description:
Linden Lab provides open source official viewers for. The viewers can be used as a basis for various modified viewers.
Second Life
Description:
Linden Lab provides open source official viewers for. The viewers can be used as a basis for various modified viewers.
 
Breaking Simulation: The Return Of The Last Robot
Description:
Recent developments in computer games and in Interactive TV have led to the creation of new gaming applications such as interactive gaming (game played by a couple of players) and more. With that in mind, these projects aim at achieving real-time interception of game figures allowing the player full visibility of the figure's behavior. The projects are in cooperation with a startup company which specializes ingaming systems.
General Augmented Reality Games
Description:
In this project we will employ the 3D camera to create Augmented Reality Games. The 3D camera can be used to translate the real world objects into computer 3D reality and to activate them according to user gestures.
 
Hand Gesture Recognition
Best lab Project of 2009
Description:
Interaction with personal computers has become one of the most common and trivial tasks in our everyday life. But despite the huge advances in computer hardware, the interaction itself has not changed much since the invention of the mouse. In this project, we develop a much more intuitive system for human-computer interaction.
Gesture Recognition
Description:
Human Computer gesture based interaction has the potential to speed-up work process, provide better user experience and simplify otherwise complicated tasks.
 
Scene Completion Using Millions of Photographs.
Description:
What can you do with million images? In this project a new image completion algorithm powered by a huge database of photographs gathered from the Web will be implemented...
Coded Aperture solution using de-convolution
Description:
Patterned mask inserted at the aperture stop of a camera preserves high frequency components and increases the sensitivity of details sharpness to defocus. This can be used to derive depth map through an estimated blurring function as well as to control the focal plane, in order to be able to focus on different objects in an image.

This project was done in cooperation with Intel.
 
Animating Animal Motion from Still
Description:
Even though the temporal information is lost, a still picture of moving animals hints at their motion. In this project, we infer motion cycle of animals from the "motion snapshots" (snapshots of different individuals) captured in a still picture.
Image Resizing
Description:
In this project student shall investigate and implement ways to resize images or elements in images while considering the content of the image, so that the content seams to remain the same size.
 
 
2007
Non Photorealistic Rendering
Description:
In this project we will investigate the ability of computer to mimic artists and to create visually appealing line drawings.
 
3D Implementation of Insect Aerodynamic Flight Simulation
Description:
There are many different sources for 3D models, among them the World Wide Web. However, there is no guarantee to their quality. In this project the students will utilize computer graphic skills and capabilities to create the ultimate butterfly model.
Real Time Game Figures Interception and Processing
Description:
Recent developments in computer games and in Interactive TV have led to the creation of new gaming applications such as interactive gaming (game played by a couple of players) and more. With that in mind, these projects aim at achieving real-time interception of game figures allowing the player full visibility of the figure's behavior. The projects are in cooperation with a startup company which specializes ingaming systems.
 
Collision Detection
Description:
In this project, we will develop an algorithm that is able to run in real-time and that is suitable for 2.5D surfaces, such as the ones created by a range camera. See the link for an example of the usage of such algorithm.
Poisson Image Editing Part 2
Description:
Image editing tasks concern global changes or local changes confined to a selection. Here the aim is to achieve local changes, restricted to a region manually selected, in a seamless and effortless manner.
 
Removing Photography Artifacts
Description:
In this project we will use a flash and no-flash (ambient) image pair to produce better flash images, Using a novel gradient projection scheme based on a gradient coherence model that allows removal of reflections and highlights from flash images.
Real Time Implementation of MPEG-2 Decoder on NVIDIA GPU
Description:
GPU stands for General-Purpose computation on GPUs. GPUs are now capable coprocessors, and their high speed makes them useful for a variety of applications. This project is held In association with: Elbit Systems.
 
Facial Animation
Description:
We have set out to create a software which will generate a movie of a person (on which we have previously gathered data our Marionette) speaking the words of another, our Actor, who moves his lips in whatever fashion he chooses. The input to the system is a video featuring a person, speaking to the camera. The output is a video of the Marionette, moving it\s lips the same way.
Underwater FireWire Camera
Description:
Raw video capture is required by researchers at the Technion for underwater imaging research. This project's goal was to design and build a custom made, compact digital video camera that would enable capturing raw video sequences in an underwater environment.
 
ACE Algorithm
Description:
This project presents a new algorithm for digital images. It uses unsupervised enhancement with simultaneous global and local effects. The algorithm is based on a computational model of some adaptation mechanisms of the human visual system.
 
2006
Real-time Shadow Mapping
Description:
In this project the student will implement an efficient real-time hardware accelerated method for rendering high-quality, antialiased shadows for complex scenes using the shadow mapping technique. The project will be done in cooperation with the electro optical visualization section in Rafael, Missile Division.
 
Fast special effects using particle systems
Description:
"Particle System" is a system that enables imitating the true natural behavior of particles. It serves users that wish to create an effect related to particles in real time with a high Frame per Second rate. Its modular structure allows for implementing effects like smoke or dust by a simple change of a set of parameters.
Synthesizing textures and images using quilting and wang tiles.
Description:
Texture synthesis is a major research topic in computer vision, image processing and computer graphics. In this project we will investigate and implement two algorithms of textures synthesis.
 
Poisson Image Editing
Description:
Image editing tasks concern global changes or local changes confined to a selection. Here the aim is to achieve local changes, restricted to a region manually selected, in a seamless and effortless manner.
Underwater Vision
Description:
In this project the students used Matlab to implement a specific algorithm suggested by Dr. Yoav Shechner for dehazing underwater pictures. The project required defining the keys and environmental conditions to ensure the algorithm's success.
 
Image Fusion
Description:
In this project we utilize a variety of pyramid methods developed for image data compression, enhancement, Analysis, and graphic. The applications implemented in this project are: Image compression, Multi-Focus composition, Boundary Detection, and Image Mosaics.
 
2005
Mesh morphing using Matlab - 2005
Description:
In this project the students will use Matlab to implement mesh metamorphosis. The implementation will be either in pure Matlab or in c/c++ with an interface to Matlab.
 
 
 
New Projects Current Projects Archive Projects
CG&M Lab    Contact Us EE Labs EE Department Technion