Person: Tatu, Aditya
Loading...
Name
Aditya Tatu
Job Title
Faculty
Email Address
Telephone
079-68261540
Birth Date
Specialization
Computer Vision, Image Processing, Pattern Recognition, Signal Processing
Abstract
Biography
Aditya Tatu received his Phd from the Department of Computer Science, University of Copenhagen, Denmark in 2010, after which he joined DAIICT, Gandhinagar. He is an Associate Professor at DAIICT since January 2019.
His research interests lie in the areas of Image, Signal and Geometry Processing, often relying on Variational methods, Differential Geometry, Linear algebra and (convex) optimization.
Research Projects
Organizational Units
Name
7 results
Search Results
Now showing 1 - 7 of 7
Publication Metadata only Efficient Filtering of Graph Based Data Using Graph Partitioning(Zenodo, 01-08-2017) Vaishnav,Nileshkumar; Tatu, Aditya; Tatu, Aditya; Tatu, Aditya; Tatu, Aditya; Tatu, Aditya; Tatu, Aditya; DA-IICT, Gandhinagar; Vaishnav,Nileshkumar (201121007)An algebraic framework for processing graph signals axiomatically designates the graph adjacency matrix as the shift operator. In this setup, we often encounter a problem wherein we know the filtered output and the filter coefficients, and need to find out the input graph signal. Solution to this problem using direct approach requires O(N3) operations, where N is the number of vertices in graph. In this paper, we adapt the spectral graph partitioning method for partitioning of graphs and use it to reduce the computational cost of the filtering problem. We use the example of denoising of the temperature data to illustrate the efficacy of the approach.Publication Metadata only Active Contour Models for Manifold Valued Image Segmentation(Springer, 01-06-2015) Bansal, Sumukh; Tatu, Aditya; Tatu, Aditya; Tatu, Aditya; Tatu, Aditya; Tatu, Aditya; Tatu, Aditya; DA-IICT, Gandhinagar; Bansal, Sumukh (201421002)Image segmentation is the process of partitioning an image into different regions or groups based on some characteristics like color, texture, motion, or shape etc. Active contours are a popular variational method for object segmentation in images, in which the user initializes a contour which evolves in order to optimize an objective function designed such that the desired object boundary is the optimal solution. Recently, imaging modalities that produce Manifold-valued images are frequently used, for example, DT-MRI images, vector fields. The traditional active contour model does not work on such images. In this paper, we generalize the active contour model to work on Manifold-valued images. As expected, our algorithm detects regions with similar Manifold values in the image. Our algorithm also produces expected results on usual gray-scale images, since these are nothing but trivial examples of Manifold-valued images. As another application of our general active contour model, we perform texture segmentation on gray-scale images by first creating an appropriate Manifold-valued image. We demonstrate segmentation results for manifold-valued images and texture images.Publication Metadata only Signal processing on graphs: structure preserving maps(IET, 01-02-2019) Vaishnav,Nileshkumar; Tatu, Aditya; Tatu, Aditya; Tatu, Aditya; Tatu, Aditya; Tatu, Aditya; Tatu, Aditya; DA-IICT, Gandhinagar; Vaishnav,Nileshkumar (201121007)Signal processing on graphs using adjacency matrix (as opposed to more traditional graph Laplacian) results in an algebraic framework for graph signals and shift invariant filters. This can be seen as an example of the algebraic signal processing theory. In this study, the authors examine the concepts of homomorphism and isomorphism between two graphs from a signal processing point of view and refer to them as GSP isomorphism and GSP homomorphism, respectively. Collectively, they refer to these concepts as structure preserving maps (SPMs). The fact that linear combination of signals and linear transforms on signals are meaningful operations has implications on the GSP isomorphism and GSP homomorphism, which diverges from the topological interpretations of the same concepts (i.e. graph isomorphism and graph homomorphism). When SPMs exist between two graphs, signals and filters can be mapped between them while preserving spectral properties. They examine conditions on adjacency matrices for such maps to exist. They also show that isospectral graphs form a special case of GSP isomorphism and that GSP isomorphism and GSP homomorphism is intrinsic to resampling and downsampling process.Publication Metadata only Affine interpolation in a lie group framework(ACM, 12-07-2019) Bansal, Sumukh; Tatu, Aditya; Tatu, Aditya; Tatu, Aditya; Tatu, Aditya; Tatu, Aditya; Tatu, Aditya; DA-IICT, Gandhinagar; Bansal, Sumukh (201421002)Affine transformations are of vital importance in many tasks pertaining to motion design and animation. Interpolation of affine transformations is non-trivial. Typically, the given affine transformation is decomposed into simpler components which are easier to interpolate. This may lead to unintuitive results, while in some cases, such solutions may not work. In this work, we propose an interpolation framework which is based on a Lie group representation of the affine transformation. The Lie group representation decomposes the given transformation into simpler and meaningful components, on which computational tools like the exponential and logarithm maps are available in closed form. Interpolation exists for all affine transformations while preserving a few characteristics of the original transformation. A detailed analysis and several experiments of the proposed framework are included.Publication Metadata only On restricting planar curve evolution to finite dimensional implicit subspaces with non-euclidean metric(01-11-2010) Tatu, Aditya; Lauze, Franois; Sommer, Stefan; Nielsen, Mads; Tatu, Aditya; Tatu, Aditya; Tatu, Aditya; Tatu, Aditya; Tatu, Aditya; DA-IICT, GandhinagarThis paper deals with restricting curve evolution to a finite and not necessarily flat space of curves, obtained as a subspace of the infinite dimensional space of planar curves endowed with the usual but weak parametrization invariant curve�L�2-metric.Publication Metadata only Image similarity based on intensity using mutual information(01-04-2013) Mistry, D; Banerjee, Asim; Tatu, Aditya; DA-IICT, GandhinagarPublication Metadata only Interval-Based Least Squares for Uncertainty-Aware Learning in Human-Centric Multimedia Systems(IEEE, 11-11-2021) Narwaria, Manish; Tatu, Aditya; DA-IICT, GandhinagarMachine learning (ML) methods are popular in several application areas of multimedia signal processing. However, most existing solutions in the said area, including the popular least squares, rely on penalizing predictions that deviate from the target ground-truth values. In other words, uncertainty in the ground-truth data is simply ignored. As a result, optimization and validation overemphasize a single-target value when, in fact, human subjects themselves did not unanimously agree to it. This leads to an unreasonable scenario where the trained model is not allowed the benefit of the doubt in terms of prediction accuracy. The problem becomes even more significant in the context of more recent human-centric and immersive multimedia systems where user feedback and interaction are influenced by higher degrees of freedom (leading to higher levels of uncertainty in the ground truth). To ameliorate this drawback, we propose an uncertainty aware loss function (referred to as�MSE?�) that explicitly accounts for data uncertainty and is useful for both optimization (training) and validation. As examples, we demonstrate the utility of the proposed method for blind estimation of perceptual quality of audiovisual signals, panoramic images, and images affected by camera-induced distortions. The experimental results support the theoretical ideas in terms of reducing prediction errors. The proposed method is also relevant in the context of more recent paradigms, such as crowdsourcing, where larger uncertainty in ground truth is expected.