Robust recognition of hand gestures in real-world applications is still an unaccomplished goal du... more Robust recognition of hand gestures in real-world applications is still an unaccomplished goal due to many remaining challenges, such as cluttered backgrounds and unconstrained environmental factors. In most existing methods in this field, hand segmentation is a critical step to reduce redundant information in the scene background, in preparation for the hand gesture recognition stage. Therefore, we propose a new two-stage convolutional neural network (CNN) architecture, called HGR-Net, where the first stage performs accurate pixel-level semantic segmentation to determine the hand regions, and the second stage identifies the hand gesture. The segmentation stage architecture is based on the combination of fully convolutional residual network and atrous spatial pyramid pooling. Although the segmentation sub-network is trained without depth information, it is significantly robust against challenges such as illumination variations and complex backgrounds. The recognition stage deploys a...
Robust recognition of hand gesture in real-world applications is still a challenging task due to ... more Robust recognition of hand gesture in real-world applications is still a challenging task due to the many aspects such as cluttered backgrounds and uncontrolled environment factors. In most existing methods hand segmentation is a primary step for hand gesture recognition, because it reduces redundant information from the image background, before passing them to the recognition stages. Therefore, in this paper we propose a two-stage deep convolutional neural network (CNN) architecture called HGR-Net, where the first stage performs accurate pixel-level semantic segmentation into hand region and the second stage identifies hand gesture style. The segmentation stage architecture is based on the combination of fully convolutional deep residual neural network and atrous spatial pyramid pooling. Although the segmentation sub-network is trained without using depth information, it is robust enough against challenging situations such as changes in the lightning and complex backgrounds. In the...
Hand gesture recognition is a challenging problem in human computer interaction. A familiar categ... more Hand gesture recognition is a challenging problem in human computer interaction. A familiar category of this problem is American sign language recognition. In this paper, we study this problem from a topological point of view. We introduce a novel topological feature to capture and represent the shape properties of a hand gesture. The method is invariant to changes in rotation, scale, noise, and articulations. Due to the lack of ASL image database with all variations and signs, we introduced a database consisting of 520 images of 26 ASL gestures, with different rotation and deformations. Experimental results show that this algorithm can achieve a higher performance in comparison with state of the art methods.
In many industrial and non-industrial applications, it is necessary to identify the largest inscr... more In many industrial and non-industrial applications, it is necessary to identify the largest inscribed rectangle in a certain shape. The problem is studied for convex and non-convex polygons. Another criterion is the direction of the rectangle: axis aligned or general. In this paper a heuristic algorithm is presented for finding the largest axis aligned inscribed rectangle in a general polygon. Comparing with stare of the art, the rectangles resulted from our algorithm have bigger area. We also proposed an approach to use the algorithm for finding a rectangle with general direction.
By an ortho-radial drawing of a graph we mean a planar drawing on concentric circles such that ea... more By an ortho-radial drawing of a graph we mean a planar drawing on concentric circles such that each edge is an alternating sequence of circular and radial segments, where a circular segment is a part of a circle and a radial segment is a part of a half-line starting at the center of the circles. Ortho-radial drawings are topologically an extension of orthogonal drawings to drawings on a cylinder. We study the relationship between ortho-radial drawings and orthogonal drawings, then we prove necessary and sufficient conditions for a path, cycle or a theta graph to have an ortho-radial drawing consistent with a C-shape (cylindrical shape) which is a specification of the direction in which each edge must be drawn. Furthermore, we present an example of a C-shape of a graph such that all of its cycles have an ortho-radial drawing but the graph itself does not have any ortho-radial drawing with this C-shape. This is in contrast to the properties of orthogonal drawings on the plane.
The concept of graph burning and burning number ($bn(G)$) of a graph G was introduced recently [1... more The concept of graph burning and burning number ($bn(G)$) of a graph G was introduced recently [1]. Graph burning models the spread of contagion (fire) in a graph in discrete time steps. $bn(G)$ is the minimum time needed to burn a graph $G$.The problem is NP-complete. In this paper, we develop first heuristics to solve the problem in general (connected) graphs. In order to test the performance of our algorithms, we applied them on some graph classes with known burning number such as theta graphs, we tested our algorithms on DIMACS and BHOSLIB that are known benchmarks for NP-hard problems in graph theory. We also improved the upper bound for burning number on general graphs in terms of their distance to cluster. Then we generated a data set of 2000 random graphs with known distance to cluster and tested our heuristics on them.
Shape recognition is the main challenging problem in computer vision. Different approaches and to... more Shape recognition is the main challenging problem in computer vision. Different approaches and tools are used to solve this problem. Most existing approaches to object recognition are based on pixels. Pixel-based methods are dependent on the geometry and nature of the pixels, so the destruction of pixels reduces their performance. In this paper, we study the ability of graphs as shape recognition. We construct a graph that captures the topological and geometrical properties of the object. Then, using the coordinate and relation of its vertices, we extract features that are robust to noise, rotation, scale variation, and articulation. To evaluate our method, we provide different comparisons with state-of-the-art results on various known benchmarks, including Kimia's, Tari56, Tetrapod, and Articulated dataset. We provide an analysis of our method against different variations. The results confirm our performance, especially against noise.
Given a graph G, a drawing of G is a function that maps each vertex v to a distinct point ( v) in... more Given a graph G, a drawing of G is a function that maps each vertex v to a distinct point ( v) in the plane and each edge uv to a simple open Jordan curve ( uv) with endpoints u and v. Graph drawing, in general, deals with automatic generation of drawings for a given graph. There are dierent standards for drawing a graph. In this paper we emphasis on combinatorial aspects of graph drawing. We present necessary and sucient combinatorial condition for an embedded graph with given shape to have orthogonal drawing, and for a digraph to have upward drawing in the plane. Then we extend the results to higher dimensions and some surfaces.
Hand Gesture Recognition (HGR) is of major importance for Human-Computer Interaction (HCI) applic... more Hand Gesture Recognition (HGR) is of major importance for Human-Computer Interaction (HCI) applications. In this paper, we present a new hand gesture recognition approach called GNG-IEMD. In this approach, first, we use a Growing Neural Gas (GNG) graph to model the image. Then we extract features from this graph. These features are not geometric or pixel-based, so do not depend on scale, rotation, and articulation. The dissimilarity between hand gestures is measured with a novel Improved Earth Mover\textquotesingle s Distance (IEMD) metric. We evaluate the performance of the proposed approach on challenging public datasets including NTU Hand Digits, HKU, HKU multi-angle, and UESTC-ASL and compare the results with state-of-the-art approaches. The experimental results demonstrate the performance of the proposed approach.
By an ortho-radial drawing of a graph we mean a planar drawing on concentric circles such that ea... more By an ortho-radial drawing of a graph we mean a planar drawing on concentric circles such that each edge is an alternating sequence of circular and radial segments, where a circular segment is a part of a circle and a radial segment is a part of a half-line starting at the center of the circles. Ortho-radial drawings are topologically an extension of orthogonal drawings to drawings on a cylinder. We study the relationship between ortho-radial drawings and orthogonal drawings, then we prove necessary and sufficient conditions for a path, cycle or a theta graph to have an ortho-radial drawing consistent with a C-shape (cylindrical shape) which is a specification of the direction in which each edge must be drawn. Furthermore, we present an example of a C-shape of a graph such that all of its cycles have an ortho-radial drawing but the graph itself does not have any ortho-radial drawing with this C-shape. This is in contrast to the properties of orthogonal drawings on the plane.
The polar diagram of a set of points in a plane and its extracted dual EDPD were recently introdu... more The polar diagram of a set of points in a plane and its extracted dual EDPD were recently introduced for static and dynamic cases. In this paper, the near-pole polar diagram NPPD for a set of points is presented. This new diagram can be considered as a generalization of the polar diagram and has applications in several communication systems and
Area compaction of an orthogonal representation H states for computing a planar drawing of H with... more Area compaction of an orthogonal representation H states for computing a planar drawing of H with small area. Patrignani [On the complexity of orthogonal compaction, in: F. Dehne, A. Gupta, J.-R. Sack, R. Tamassia (Eds.), Algorithms and Data Structures, Proceedings of ...
Robust recognition of hand gestures in real-world applications is still an unaccomplished goal du... more Robust recognition of hand gestures in real-world applications is still an unaccomplished goal due to many remaining challenges, such as cluttered backgrounds and unconstrained environmental factors. In most existing methods in this field, hand segmentation is a critical step to reduce redundant information in the scene background, in preparation for the hand gesture recognition stage. Therefore, we propose a new two-stage convolutional neural network (CNN) architecture, called HGR-Net, where the first stage performs accurate pixel-level semantic segmentation to determine the hand regions, and the second stage identifies the hand gesture. The segmentation stage architecture is based on the combination of fully convolutional residual network and atrous spatial pyramid pooling. Although the segmentation sub-network is trained without depth information, it is significantly robust against challenges such as illumination variations and complex backgrounds. The recognition stage deploys a...
Robust recognition of hand gesture in real-world applications is still a challenging task due to ... more Robust recognition of hand gesture in real-world applications is still a challenging task due to the many aspects such as cluttered backgrounds and uncontrolled environment factors. In most existing methods hand segmentation is a primary step for hand gesture recognition, because it reduces redundant information from the image background, before passing them to the recognition stages. Therefore, in this paper we propose a two-stage deep convolutional neural network (CNN) architecture called HGR-Net, where the first stage performs accurate pixel-level semantic segmentation into hand region and the second stage identifies hand gesture style. The segmentation stage architecture is based on the combination of fully convolutional deep residual neural network and atrous spatial pyramid pooling. Although the segmentation sub-network is trained without using depth information, it is robust enough against challenging situations such as changes in the lightning and complex backgrounds. In the...
Hand gesture recognition is a challenging problem in human computer interaction. A familiar categ... more Hand gesture recognition is a challenging problem in human computer interaction. A familiar category of this problem is American sign language recognition. In this paper, we study this problem from a topological point of view. We introduce a novel topological feature to capture and represent the shape properties of a hand gesture. The method is invariant to changes in rotation, scale, noise, and articulations. Due to the lack of ASL image database with all variations and signs, we introduced a database consisting of 520 images of 26 ASL gestures, with different rotation and deformations. Experimental results show that this algorithm can achieve a higher performance in comparison with state of the art methods.
In many industrial and non-industrial applications, it is necessary to identify the largest inscr... more In many industrial and non-industrial applications, it is necessary to identify the largest inscribed rectangle in a certain shape. The problem is studied for convex and non-convex polygons. Another criterion is the direction of the rectangle: axis aligned or general. In this paper a heuristic algorithm is presented for finding the largest axis aligned inscribed rectangle in a general polygon. Comparing with stare of the art, the rectangles resulted from our algorithm have bigger area. We also proposed an approach to use the algorithm for finding a rectangle with general direction.
By an ortho-radial drawing of a graph we mean a planar drawing on concentric circles such that ea... more By an ortho-radial drawing of a graph we mean a planar drawing on concentric circles such that each edge is an alternating sequence of circular and radial segments, where a circular segment is a part of a circle and a radial segment is a part of a half-line starting at the center of the circles. Ortho-radial drawings are topologically an extension of orthogonal drawings to drawings on a cylinder. We study the relationship between ortho-radial drawings and orthogonal drawings, then we prove necessary and sufficient conditions for a path, cycle or a theta graph to have an ortho-radial drawing consistent with a C-shape (cylindrical shape) which is a specification of the direction in which each edge must be drawn. Furthermore, we present an example of a C-shape of a graph such that all of its cycles have an ortho-radial drawing but the graph itself does not have any ortho-radial drawing with this C-shape. This is in contrast to the properties of orthogonal drawings on the plane.
The concept of graph burning and burning number ($bn(G)$) of a graph G was introduced recently [1... more The concept of graph burning and burning number ($bn(G)$) of a graph G was introduced recently [1]. Graph burning models the spread of contagion (fire) in a graph in discrete time steps. $bn(G)$ is the minimum time needed to burn a graph $G$.The problem is NP-complete. In this paper, we develop first heuristics to solve the problem in general (connected) graphs. In order to test the performance of our algorithms, we applied them on some graph classes with known burning number such as theta graphs, we tested our algorithms on DIMACS and BHOSLIB that are known benchmarks for NP-hard problems in graph theory. We also improved the upper bound for burning number on general graphs in terms of their distance to cluster. Then we generated a data set of 2000 random graphs with known distance to cluster and tested our heuristics on them.
Shape recognition is the main challenging problem in computer vision. Different approaches and to... more Shape recognition is the main challenging problem in computer vision. Different approaches and tools are used to solve this problem. Most existing approaches to object recognition are based on pixels. Pixel-based methods are dependent on the geometry and nature of the pixels, so the destruction of pixels reduces their performance. In this paper, we study the ability of graphs as shape recognition. We construct a graph that captures the topological and geometrical properties of the object. Then, using the coordinate and relation of its vertices, we extract features that are robust to noise, rotation, scale variation, and articulation. To evaluate our method, we provide different comparisons with state-of-the-art results on various known benchmarks, including Kimia's, Tari56, Tetrapod, and Articulated dataset. We provide an analysis of our method against different variations. The results confirm our performance, especially against noise.
Given a graph G, a drawing of G is a function that maps each vertex v to a distinct point ( v) in... more Given a graph G, a drawing of G is a function that maps each vertex v to a distinct point ( v) in the plane and each edge uv to a simple open Jordan curve ( uv) with endpoints u and v. Graph drawing, in general, deals with automatic generation of drawings for a given graph. There are dierent standards for drawing a graph. In this paper we emphasis on combinatorial aspects of graph drawing. We present necessary and sucient combinatorial condition for an embedded graph with given shape to have orthogonal drawing, and for a digraph to have upward drawing in the plane. Then we extend the results to higher dimensions and some surfaces.
Hand Gesture Recognition (HGR) is of major importance for Human-Computer Interaction (HCI) applic... more Hand Gesture Recognition (HGR) is of major importance for Human-Computer Interaction (HCI) applications. In this paper, we present a new hand gesture recognition approach called GNG-IEMD. In this approach, first, we use a Growing Neural Gas (GNG) graph to model the image. Then we extract features from this graph. These features are not geometric or pixel-based, so do not depend on scale, rotation, and articulation. The dissimilarity between hand gestures is measured with a novel Improved Earth Mover\textquotesingle s Distance (IEMD) metric. We evaluate the performance of the proposed approach on challenging public datasets including NTU Hand Digits, HKU, HKU multi-angle, and UESTC-ASL and compare the results with state-of-the-art approaches. The experimental results demonstrate the performance of the proposed approach.
By an ortho-radial drawing of a graph we mean a planar drawing on concentric circles such that ea... more By an ortho-radial drawing of a graph we mean a planar drawing on concentric circles such that each edge is an alternating sequence of circular and radial segments, where a circular segment is a part of a circle and a radial segment is a part of a half-line starting at the center of the circles. Ortho-radial drawings are topologically an extension of orthogonal drawings to drawings on a cylinder. We study the relationship between ortho-radial drawings and orthogonal drawings, then we prove necessary and sufficient conditions for a path, cycle or a theta graph to have an ortho-radial drawing consistent with a C-shape (cylindrical shape) which is a specification of the direction in which each edge must be drawn. Furthermore, we present an example of a C-shape of a graph such that all of its cycles have an ortho-radial drawing but the graph itself does not have any ortho-radial drawing with this C-shape. This is in contrast to the properties of orthogonal drawings on the plane.
The polar diagram of a set of points in a plane and its extracted dual EDPD were recently introdu... more The polar diagram of a set of points in a plane and its extracted dual EDPD were recently introduced for static and dynamic cases. In this paper, the near-pole polar diagram NPPD for a set of points is presented. This new diagram can be considered as a generalization of the polar diagram and has applications in several communication systems and
Area compaction of an orthogonal representation H states for computing a planar drawing of H with... more Area compaction of an orthogonal representation H states for computing a planar drawing of H with small area. Patrignani [On the complexity of orthogonal compaction, in: F. Dehne, A. Gupta, J.-R. Sack, R. Tamassia (Eds.), Algorithms and Data Structures, Proceedings of ...
Uploads
Papers by Maryam Tahmasbi