Welcome to the Geometrical Image Processing Lab (GIP)!
GIP was founded in 1998 by Prof. Ron Kimmel
Aims: To conduct theoretical and applied research in geometrical image processing, three-dimensional data analysis, image and video manipulation using dictionaries and sparse representations, to promote the research of faculty members.
Challenges: Create a world excellence center in the fields of geometrical image processing, three-dimensional analysis of geometric structures, and model based image and video processing and analysis.
15.06.17 Invitation to Alon Steren's lecture
Time and Place: 21/06/2017, 14:30, Taub 601
Speaker: Alon Shtern
Title: Shape Correspondence using Spectral Methods and Deep Learning
Supervisor: Prof. Ron Kimmel
The interest in acquiring and analyzing the geometry of the world is ever increasing, fueling a wide range of computer vision algorithms in the field of geometry processing. Spectral analysis has become key component in many applications involving non-rigid shapes modeled as two-dimensional surfaces, and recently, convolutional neural networks have shown remarkable success in a variety of computer vision tasks. We designed a set of methods and tools that use these paradigms for applications such as shape correspondence, nonrigid deformations, and volumetric optical flow. In this talk we will present three different ways to infer point-to-point correspondences between deformable shapes, which is a fundamental operation in the field of geometry processing.
A well-established approach to address the non-rigid shape correspondence problem is to define a measure of dissimilarity between the shapes. One way for measuring distance between two non-rigid shapes is to embed their two-dimensional surfaces into some common Euclidean space, defining the comparison task as a problem of rigid matching in that space. In the first part of this talk we review a novel spectral embedding, named the "Spectral Gradient Fields Embedding", which exploits the local interactions between the eigenfunctions of the Laplace-Beltrami operator and the extrinsic geometry of the surface.
Next, we analyze the applicability of the spectral kernel distance, as a measure of dissimilarity between surfaces, for solving the shape matching problem. To align the spectral kernels, we developed the Iterative Closest Spectral Kernel Maps (ICSKM) algorithm. ICSKM extends the Iterative Closest Point (ICP) algorithm to the class of deformable shapes. Instead of aligning the shapes in the three dimensional Euclidean domain, this method estimates the transformation that best fits the embeddings of the shapes into the spectral domain.
Volumetric optical flow is a different way to address the matching problem of a three-dimensional dynamic scene. In the last part of the talk we introduce a multi-scale optical flow based deep learning architecture for predicting the next frame of a sequence of volumetric images. The fully differentiable model consists of specific crafted modules that are trained on small patches in an unsupervised manner. The approach, called "V-Flow", is useful for analyzing the temporal dynamics of three-dimensional images in applications that involve, for example, motion of viscous fluid substances or volumetric medical imaging.
13.06.17 Aaron Wetzler's PhD Seminar on Computational Geometric Vision
Who: Aaron Wetzler (EE/CS-Technion)
Time and Place: Thursday 22/06, 11.30 in Taub 337
Title:PhD Seminar: Computational Geometric Vision
By combining geometric principles for shape analysis with modern sensing techniques, large-scale datasets and powerful
computational architectures we show various new ways of enabling computers to better perceive, interpret and comprehend
the geometry of the world around them. Specifically we explore the topics of reconstruction, filtering and semantic
processing within the context of computational geometric vision.
Reconstruction - We start by discussing the problem of sensing and reconstructing three-dimensional geometry. We develop a method of performing efficient photometric stereo which can model non-linear and near-source lighting setups while avoiding directly computing normal fields. We then look into to two alternative approaches for reconstructing geometry on a smartphone using projected light.
Filtering - Image and shape data which is obtained through modern cameras and sensors is typically noisy. We contribute a new patch based data denoising framework called the Beltrami Patch filter for denoising grayscale, color images and extend it to depth fields and 3D meshes. We then modify the approach by reformulating the differential operators used as trainable kernels in a deep neural network and unrolling the update step through time. We demonstrate state of the art results and highlight the fact that other PDE based methods could take advantage of the same basic idea.
Semantic processing - In the last part of this talk we discuss the problem of localizing and identifying self similar geometric objects in a complex visual space. We specifically focus on the problem of identifying fingertips on articulating hands observed by depth cameras. We describe how we used high accuracy magnetic sensors to annotate large quantities of training data for both front facing and ego-centric hand motion. To perform learning efficiently we turn to random forests and contribute a new approach for efficiently mapping the training of a random decision tree on billions of training samples with trillions of features to a single multi-GPU computing node. Similarly for inference we describe a novel pipelined FPGA hardware implementation.
PhD work supervised by Prof. Ron Kimmel
13.06.16 Dan Raviv gave a superb talk in SIAM
SIAG/Imaging Science Best Paper Prize Lecture - Scale Invariant Geometry for Nonrigid Shapes
31.05.16 Matan Sela won 1st prize in the CS Sixth Research Day, 2016
The CS Department held its sixth Research Day on Tuesday, May 31 2016. A substantial number of posters and presentations (29) were exhibited by excellent M.Sc. and Ph.D. students before multitude of guests from the Technion and the industry.
08.05.16 Gil Ben-Shachar and Elad Richardson won the the Amdocs Best Project Contest 2016
The project "Automatic generation of Planar marionettes from frontal images" was supervised by Anastasia Dubrovina-Karni and Aaron Wetzler.
Congratulation to the winners!!!
07.10.15 Ron slosberg winning the final prize
Congratulation to Ron for his achievement.
17.09.15 Prof. MIchael Elad Among 10 Israeli Highly Cited Researchers 2015
Congratulation to Prof. MIchael Elad who was named one of the top ten most Highly Cited Israeli Researchers in 2015 as published by Thomson Reuters rank of the world’s most influential scientific minds.
ברכות לאלעד לרגל זכייתו בפרוייקט מצטיין פקולטי בתחרות אמדוקס לשנת 2015
ברכות לאלעד לגל זכייתו בפרוייקט פקולטי בתחרות אמדוקס לשנת 2015.
באותה נשימה אני מודה למנחים המצויינים של הפרוייקט אנאסטסיה ואהרון ישר כוח.
את הפרוייקט ניתן למצוא בלינק הבא:
קבוצת סטודנטים מהמעבדה זכתה במקום החמישי בתחרות AUVSI 2014
קבוצת סטודנטים מהפקולטה שלנו במסגרת פרוייקט במעבדה לעיבוד תמונה גיאומטרי(GIP)ובשיתוף עם סטודנטים מהפקולטה לאוירונאוטיקה השתתפה לראשונה בתחרות המטוס האוטונומי של ארגון AUVSI בתחרות השתתפו 44 קבוצות מרחבי העולם אשר 33 קבוצות מתוכן הצליחו לעמוד בדרישות התחרות. קבוצת הסטודנטים ממדעי המחשב כללה את ליאת כץ, לי-טל, קופרמן ,נדב אופיר, זיו נחום, אלי חיון ואיתי גיא בהנחייתם של אהרון ווצלר ומהנדס המעבדה ירון חונן. קבוצת הסטודנטים ממדעי המחשב בנתה מערכת לזיהוי מטרות בזמן טיסה שכללה מערכת מוטסת המשדרת לתחנת קרקעית, זכתה במקום החמישי. התחרות נערכה במרילנד, ארה"ב, תחת הנחייתו של דרור ארצי,http://www.technion.ac.il/2014/11/צמד-בשחקים/
כתבה באתר israeldefense:
15.09.14 New projects for upcoming semester
Please see (in the proposed project) new projects for the upcoming semester.
20.07.14 Prof. Michael Elad on Thomson Reuters List of Most Influential Scientists in their Respective Areas
The list, published by Thomson Reuters company, was based on the analysis of citation data for the last 11 years, emphasizing the last two years. In particular, the company defined "highly-cited" papers as papers whose citation count is in the top 1% in their areas. Altogether, 21 broad areas were defined (Biology Biochemistry, Engineering, Computer Science, etc.).
16.06.14 Congratulations to Prof. Ron Kimmel
Professor Ron Kimmel won the Cooper Award for Excellence in Research for 2014.
01.05.14 Congratulations to Shahar and Omri
GIP students shahar Sagiv and Omri Panziel with their Puzel Bingo Project (supervised by Yaron Honen and Itay Dabran - LCCN) were in the team that won the 3rd prize of The 2nd Intel Challenge contest.
21.04.14 Cool Technologies from the Technion: 3D Caricature Music Clip
From the Technion Youtube - See the clip 3D Caricature Music Clip
03.04.14 3D Photographing Work Shop in Alberta , Canada
Yaron Honen, facilitated a Women’s and Children’s Health Research Institute (WCHRI) workshop on the uses of 3D photography in Alberta University, Canada. Read more...
30.03.14 Cool Technologies from the Technion: Technion Plays PONG to Go
From the Technion Youtube - See the clip PONG to Go
30.03.14 Congratulations to Vered and Nurit
Vered Cohen and Nurit Schwevsky (supervised by Aaron Wetzler): "Real Time 3D Surface Reconstruction" was one of the winning projects of the 2013 Amdocs Best Project Contest.
23.03.14 Cool Technologies from the Technion: Geometric Image Processing Lab 3D Scanner
From the Technion Youtube - See the clip Geometric Image Processing Lab 3D Scanner