Differences

This shows you the differences between two versions of the page.

--- start [2013/12/24 18:07]
suyash
+++ start [2022/09/04 17:17]
sharat [Welcome]
@@ Line 1: / Line 1: @@
-====== Welcome to the ViGIL Community!!! ======
+{{ :logo2020.png?direct&500 |}}
-We are a bunch of folks dabbling with images, and more recently media of all kind. The whole idea is that it doesn't make sense to do computing, if you don't do **visual** computing.
+===== Welcome =====
-Well, that said, this image will give you the general idea.
-{{flow.png|}}
+We are a bunch of folks in the [[ https://www.cse.iitb.ac.in | Department of Computer Science and Engineering ]] at the [[ https://www.iitb.ac.in | Indian Institute of Technology (IIT) Bombay ]].
+We dabble with pictures, including technology involving image creation, acquisition, representation, visualization, manipulation, processing, and analysis. This involves following principles in computing, mathematics, and science.
 ===== News =====
-===== Latest Papers =====
+Excellence in Teaching 2022: Parag has been selected to receive the Institute award. Congratulations!
-|{{hitpoint.png|}}| ** {{ncvpripg-2010-paperid-87.pdf | Vision Assisted Safety Enhanced Shooting Range Simulator}} \\  ([[http://www.cse.iitb.ac.in/~sladha | Shamsuddin N. Ladha]], [[http://www.cse.iitb.ac.in/~sharat | Sharat Chandran]], [[http://users.monash.edu.au/~ksmiles/ | Kate Smith-Miles]]) \\ accepted at [[http://conf.lnmiit.ac.in/ncvpripg | National Conference on Computer Vision, Pattern Recognition, Image Processing and Graphics, 2010]]** \\ Traditional shooting range simulators involve large investment in real estate, and other infrastructural resources. This paper describes a novel application of projector-camera systems to create a virtual shooting range. Projector-based display systems offer an attractive combination of dense pixels over large regions. When coupled with a camera, the feedback system enables the display to be adjusted to environments enabling a variety of applications. We analyse two different implementation approaches for the simulator. As compared to a traditional shooting range, our simulator is inexpensive, safe and eco-friendly. |
-|{{http://www.cse.iitb.ac.in/~ajjoshi/randF1_new_small.png}}| ** [[http://www.cse.iitb.ac.in/~ajjoshi/hybridSVM_BIBM09.pdf | Hybrid SVM for Multiclass Arrhythmia Classification]] \\  ([[http://www.cse.iitb.ac.in/~ajjoshi | Aniruddha J. Joshi]], [[http://www.cse.iitb.ac.in/~sharat | Sharat Chandran]], Dr. V.K. Jayaraman, Dr. B.D. Kulkarni) \\ accepted at [[http://www.ittc.ku.edu/bioinformatics/BIBM09/home.php|IEEE International Conference on Bioinformatics & Biomedicine (BIBM) , 2009]]** \\ We propose a Holder-SVM detection algorithm using a novel hybrid arrangement of binary and multiclass SVMs designed to take care of class imbalance rampant in biomedical signals. As a result, we significantly reduce the number of false negatives – patients falsely classified as normal. We used the MIT-BIH Arrhythmia database for seven different arrhythmias. We compare our hybrid SVM with a suitable conventional SVM, and show better results. We also use the new arrangement for features proposed earlier, and demonstrate the gain in accuracy. Our concept of hybrid SVM is applicable to a wide variety of multiclass classification problems. |
-|{{Ocatma.png|}}| ** Hierarchical Normalized Cuts: Unsupervised Segmentation of
-Vascular Biomarkers from Ovarian Cancer Tissue Microarrays \\  (Andrew Janowczyk, [[http://www.cse.iitb.ac.in/~sharat | Sharat Chandran]], Rajendra Singh, Dimitra Sasaroli,
-George Coukos, Michael D. Feldman, and [[http://lcib.rutgers.edu/~anantm/|Anant Madabhushi]]) accepted at [[http://www.miccai2009.org|Medical Image Computing and Computer Assisted Intervention Society (MICCAI) , 2009]]** \\ Research has shown that tumor vascular markers (TVMs) may serve as potential OCa biomarkers for prognosis prediction. The ability to quickly and quantitatively estimate vascular stained regions may yield an image based metric linked to disease survival and outcome. In this paper, we present a general, robust and efficient unsupervised segmentation algorithm, termed Hierarchical Normalized Cuts (HNCut), and show its application in precisely quantifying the presence and extent of a TVM on OCa TMAs. The strength of HNCut is in the use of a hierarchically represented data structure that bridges the mean shift (MS) and the normalized cuts (NCut) algorithms. This allows HNCut to efficiently traverse a pyramid of the input image at various color resolutions, efficiently and accurately segmenting the object class of interest (in this case ESM-1 vascular stained regions) by simply annotating half a dozen pixels belonging to the target class. |
-|{{vigil:research:factintution.png|}}| **[[http://www.cse.iitb.ac.in/appu/papers/icpr08.pdf|Manifold Optimisation for Motion Factorisation]] \\  ([[http://www.cse.iitb.ac.in/appu/|Appu Shaji]] , [[http://www.cse.iitb.ac.in/~sharat | Sharat Chandran]] and [[http://www.cs.adelaide.edu.au/~dsuter/|David Suter]]) accepted at [[http://www.icpr2008.org|International Conference on Pattern Recognition , 2008]]** \\ This paper presents a novel formulation for the popular factorisation based solution for Structure from Motion. Since our measurement matrices are populated with incomplete and inaccurate data, SVD based total least squares solution are less than appropriate. Instead, we approach the problem as a non-linear unconstrained minimisation problem on the product manifold of the Special Euclidean Group ($SE_3$). The restriction of the domain of optimisation to the $SE_3$ product manifold not only implies that each intermediate solution is a plausible object motion, but also ensures better intrinsic stability for the minimisation algorithm. We compare our method with existing state of art, and show that our algorithm exhibits superior performance.|
-|**[[http://www.cse.iitb.ac.in/~ajjoshi/icpr08_prv.pdf|Arterial Pulse Rate Variability Analysis for Diagnoses]] \\  ([[http://www.cse.iitb.ac.in/~ajjoshi/|Aniruddha Joshi]] , [[http://www.cse.iitb.ac.in/~sharat | Sharat Chandran]] and V. K. Jayaraman and B. D. Kulkarni) \\ accepted at [[http://www.icpr2008.org|International Conference on Pattern Recognition , 2008]]** \\ From the Ayurvedic practitioners point of view, the cycle-to-cycle variations are sometimes more important than the average values. Human pulse rate is affected by a variety of physiological parameters and other important factors including age, sleepwake cycle, disease present, and gender. Peak-to-peak distance varies and the collection of such consecutive intervals to form arterial pulse intervals (API) time series. We introduced PRV analysis on the arterial pulse intervals by computing several time-domain (e.g. SDAPP, PP50 count), frequency-domain (e.g. power, LF/HF), and nonlinear measures (e.g. SD1, SD1/SD2), which show variations with respect to age, and the presence and absence of disorders.|
-|**[[http://www.cse.iitb.ac.in/~ajjoshi/icpr08_multifractal.pdf|Multifractality In Arterial Pulse]] \\  ([[http://www.cse.iitb.ac.in/~ajjoshi/|Aniruddha Joshi]] , [[http://www.cse.iitb.ac.in/~sharat | Sharat Chandran]] and V. K. Jayaraman and B. D. Kulkarni) \\ accepted at [[http://www.icpr2008.org|International Conference on Pattern Recognition , 2008]]** \\ Extensive research has been done to show that heartbeats are composed of the interaction of many physiological components operating on different time scales, with nonlinear and self-regulating nature. We showed the presence of scaling (self-similarity) in inhomogeneous, nonstationary pulse beats using the detrended fluctuation analysis (DFA) through a linear relationship on a log-log graph. These fractal structures were then further established by self-affine cascades of beat-tobeat fluctuations revealed by wavelet decomposition at different time scales. Finally, We showed that our multifractal spectrums vary for pulses from three age-groups and two disorders}|
-|{{vigil:research:facerecon001.png?100|}}| **[[http://www.cse.iitb.ac.in/appu/papers/nordia08.pdf|Riemannian Manifold Optimisation for Non-rigid Structure from Motion]] \\ ([[http://www.cse.iitb.ac.in/appu/|Appu Shaji]] and [[http://www.cse.iitb.ac.in/~sharat | Sharat Chandran]]) accepted at [[http://tosca.cs.technion.ac.il/nordia08/|NORDIA, CVPR workshop]]** \\ This paper address the problem of automatically extracting the 3D configurations of deformable objects from 2D features. Our focus in this work is to build on the observation that the subspace spanned by the motion parameters is a subset of a smooth manifold, and therefore we hunt for the solution in this space, rather than use heuristics (as previously attempted earlier). We succeed in this by attaching a canonical Riemannian metric, and using a variant of the non-rigid factorisation algorithm for Structure from Motion. We qualitatively and quantitatively show that our algorithm produces better results when compared to the state of art.|
+Excellence in Teaching 2019: Ajit Rajwade has been selected to receive the Departmental award. Congratulations!
+Excellence in Research: Rohit Kumar Jena (BTech Thesis Project) Congratulations to Rohit and Prof. Suyash.
-|{{vigil:research:torus.png|}}| **{{vigil:research:paper.pdf|Fast Color-Space decomposition based Environment Matting}} \\ ([[http://www.cse.iitb.ac.in/~biswarup/|Biswarup Choudhury]] and [[http://www.cse.iitb.ac.in/~deepali|Deepali Singla]]) accepted at [[http://graphics.cs.williams.edu/i3d08/|i3D 2008, Redwood City, USA]]**\\ In the literature, environment matting (EM) refers to the complex process of discovering how light in an environment interacts with an object; notably, it may transfer through a transparent object, and undergo scattering. Modeling the 3D geometry, and the index of refraction, of non-uniformly optically active substances is intractable; therefore image-based frameworks are useful. The most convincing techniques use a large number of (monochrome, or two-tone) probing images to extract the matte. In this paper, we provide an efﬁcient EM technique, for purely refractive/reﬂective objects, which uses multiple colors as cues, and use a holistic color cube as the environment (instead of repeatedly solving the problem for ﬁve sidedrops and one backdrop.)\\ \\ [[http://www.cse.iitb.ac.in/graphics/vigil_media/bis_i3d_poster_small.jpg|Poster]]\\ {{vigil:research:I3D2008_posters_ffwd.ppt|Presentation}}\\ [[http://www.cse.iitb.ac.in/graphics/vigil_media/bis_i3d_videos|Videos]]|
+Best Paper Award finalist at IEEE ISBI 2019
+A paper by Uddeshya Upadhyay was selected as a finalist for the Best Paper Award at the IEEE International Symposium on Biomedical Imaging (ISBI) 2019.  Congrats to Uddeshya and Prof. Suyash.
+Title: Robust Super-Resolution Gan, with Manifold-Based and Perception Loss
+More details at https://biomedicalimaging.org/2019/isbi-2019-awards/
+Pratik Kalshetti's poster at SIGGRAPH 2019 won the third place in the ACM Student Research Competition in the graduate student category. The work presented in the poster is a part of Pratik's Ph.D. on real-time hand shape and pose estimation. Pratik won a bronze medal and a cash award of USD200. Congrats to Prof. Parag and Pratik
-|{{vigil:research:dragon.jpg|}}| **{{vigil:research:i3d08paper.pdf|Fast, Parallel, GPU-based Construction of Space Filling Curves and Octrees}}\\ ([[http://www.cse.iitb.ac.in/~rhushabh|Rhushabh Goradia]] and [[http://www.cse.iitb.ac.in/~prekshu|Prekshu Ajmera]]) accepted at [[http://graphics.cs.williams.edu/i3d08/|i3D 2008, Redwood City, USA]]**\\ Space Filling Curves (SFC) are particularly useful in linearization of data living in two and three dimensional spaces and have been used in a number of applications in scientific computing, and visualization. Interestingly, octrees, another versatile data structure in computer graphics, can be viewed as multiple SFCs at varying resolutions, albeit with parent-child relationship. In this paper we provide a parallel implementation of SFCs and octrees on GPUs that rely on algorithms designed to minimize or eliminate communications.\\ \\ {{vigil:research:i3d08poster.pdf|Poster}}\\ {{vigil:research:i3d2008_posters_ffwd.ppt|Presentation}}|