Abstract
An efficient and accurate shape detection model plays a major role in many research areas. With the emergence of more complex shapes in real-life applications, shape recognition models need to capture the structure with more effective features to achieve high accuracy rates for shape recognition. This article presents a new method for 2D/3D shape recognition based on graph spectral domain handcrafted features, which are formulated by exploiting both an outline and a skeleton shape through the global outline and internal details. A fully connected graph is generated over the shape outline to capture the global outline representation while a hierarchically clustered graph with adaptive connectivity is formed on the skeleton to capture the structural descriptions of the shape. We demonstrate the ability of the Fiedler vector to provide the graph partitioning of the skeleton graph. The performance evaluation demonstrates the efficiency of the proposed method compared to state-of-the-art studies with increments of 4.09%, 2.2%, and 14.02% for 2D static hand gestures, 2D shapes, and 3D shapes, respectively.
- [1] . 2018. Extracting the principal shape components via convex programming. IEEE Trans. Image Process. 27, 7 (2018), 3513–3528.Google Scholar
Cross Ref
- [2] . 2015. Improving skeletal shape abstraction using multiple optimal solutions. Pattern Recogn. 48, 11 (2015), 3504–3515.Google Scholar
Digital Library
- [3] . 2019. Graph spectral domain feature learning with application to in-air hand-drawn number and shape recognition. IEEE Access 7 (2019), 159661–159673.Google Scholar
Cross Ref
- [4] . 2019. Graph spectral domain features for static hand gesture recognition. In Proceedings of the European Signal Processing Conference (EUSIPCO’19). 1–5.Google Scholar
Cross Ref
- [5] . 2020. AGSF: Adaptive graph formulation and handcrafted graph spectral features for shape representation. IEEE Access 8, 1 (2020), 1–13.Google Scholar
- [6] . 2014. Static hand gesture recognition using discriminative 2D Zernike moments. In Proceedings of the Region 10 Conference (TENCON’14). IEEE, 1–5.Google Scholar
Cross Ref
- [7] . 2013. Kinect depth stream pre-processing for hand gesture recognition. In Proceedings of the International Conference on Image Processing. IEEE, 3735–3739. Google Scholar
Cross Ref
- [8] . 2008. Disconnected skeleton: Shape at its absolute scale. IEEE Trans. Pattern Anal. Mach. Intell. 30, 12 (2008), 2188–2203.Google Scholar
Digital Library
- [9] . 2019. Skeletal descriptions of shape provide unique perceptual information for object recognition. Sci. Rep. 9 (2019), 9359.Google Scholar
Cross Ref
- [10] . 2007. Skeleton pruning by contour partitioning with discrete curve evolution. IEEE Trans. Pattern Anal. Mach. Intell. 29, 3 (2007).Google Scholar
Digital Library
- [11] . 2009. Integrating contour and skeleton for shape classification. In Proceedings of the International Conference on Computer Vision Workshops (ICCV’09). IEEE, 360–367.Google Scholar
Cross Ref
- [12] . 2018. Abstract shape representation in human visual perception. J. Exper. Psychol.: Gen. 147, 9 (2018), 1295.Google Scholar
Cross Ref
- [13] . 2011. A new 2D static hand gesture colour image dataset for ASL gestures. Res. Lett. Info. Math. Sci. 15 (2011), 12–20.Google Scholar
- [14] . 2004. Underlying principles of visual shape selectivity in posterior inferotemporal cortex. Nature Neurosci. 7, 8 (2004), 880.Google Scholar
Cross Ref
- [15] . 2002. Alignment using spectral clusters. In Proceedings of the British Machine Vision Conference (BMVC’02). 1–10.Google Scholar
Cross Ref
- [16] . 2007. A new descriptor for 2D depth image indexing and 3D model retrieval. In Proceedings of the International Conference on Image Processing, Vol. 6. IEEE, VI–373.Google Scholar
- [17] . 2009. A benchmark for 3D mesh segmentation. ACM Trans. Graph. 28 (2009), 73.Google Scholar
Digital Library
- [18] . 2007. Transformation of shape information in the ventral pathway. Curr. Opin. Neurobiol. 17, 2 (2007), 140–147.Google Scholar
Cross Ref
- [19] . 2007. Balanced graph matching. Adv. Neural Info. Process. Syst. 19 (2007), 313.Google Scholar
- [20] . 2018. Characterizing 3D shapes: A complex network-based approach. In Proceedings of the European Signal Processing Conference (EUSIPCO’18). 608–612.Google Scholar
- [21] . 2009. Skeletal shape abstraction from examples. IEEE Trans. Pattern Anal. Mach. Intell. 31, 5 (2009), 944–952.Google Scholar
Digital Library
- [22] . 2018. Monocular RGB hand pose inference from unsupervised refinable nets. In Proceedings of the the IEEE Conference on Computer Vision and Pattern Recognition Workshops. 1075–1085.Google Scholar
Cross Ref
- [23] . 2016. Fifty years of graph matching, network alignment and network comparison. Info. Sci. 346 (2016), 180–197.Google Scholar
Digital Library
- [24] . 2004. Efficient graph-based image segmentation. J. Comput. Vision 59, 2 (2004), 167–181.Google Scholar
- [25] . 1975. A property of eigenvectors of nonnegative symmetric matrices and its application to graph theory. Czech. Math. J. 25, 4 (1975), 619–633.Google Scholar
Cross Ref
- [26] . 2016. A comprehensive performance evaluation of 3D local feature descriptors. Int. J. Comput. Vision 116, 1 (2016), 66–89.Google Scholar
Digital Library
- [27] . 2017. BoSCC: Bag of spatial context correlations for spatially enhanced 3D shape representation. IEEE Trans. Image Process. 26, 8 (2017), 3707–3720.Google Scholar
Digital Library
- [28] . 1995. Polyhedral object recognition by indexing. Pattern Recogn. 28, 12 (1995), 1855–1870.Google Scholar
Digital Library
- [29] . 2016. Circle views signature: A novel shape representation for shape recognition and retrieval. Can. J. Electric. Comput. Eng. 39, 4 (2016), 274–282.Google Scholar
Cross Ref
- [30] . 2000. Shape descriptors for non-rigid shapes with a single closed contour. In Proceedings of the the Computer Vision and Pattern Recognition (CVPR’00), Vol. 1. IEEE, 424–429.Google Scholar
Cross Ref
- [31] . 2012. Combination of bag-of-words descriptors for robust partial shape retrieval. Visual Computer 28, 9 (2012), 931–942.Google Scholar
Digital Library
- [32] . 2005. A spectral technique for correspondence problems using pairwise constraints. In Proceedings of the International Conference on Computer Vision (ICCV’05). 1482–1489.Google Scholar
Digital Library
- [33] . 2014. Human body segmentation via data-driven graph cut. IEEE Trans. Cybernet. 44, 11 (2014), 2099–2108.Google Scholar
Cross Ref
- [34] . 2018. Pointcnn: Convolution on x-transformed points. In Advances in Neural Information Processing Systems. MIT Press, 820–830.Google Scholar
- [35] . 2020. A benchmark dataset and comparison study for multi-modal human action analytics. ACM Trans. Multimedia Comput. Commun. Appl. 16, 2 (2020), 1–24.Google Scholar
Digital Library
- [36] . 2015. A computational model of the short-cut rule for 2D shape decomposition. IEEE Trans. Image Process. 24, 1 (2015), 273–283.Google Scholar
Cross Ref
- [37] . 2020. FIN: Feature integrated network for object detection. ACM Trans. Multimedia Comput. Commun. Appl. 16, 2 (2020), 1–18.Google Scholar
Digital Library
- [38] . 1991. A neural-gas network learns topologies. In Proceedings of the International Conference on Artificial Neural Networks. 397–402.Google Scholar
- [39] . 2017. Spectral shape classification: A deep learning approach. J. Visual Commun. Image Represent. 43 (2017), 198–211.Google Scholar
Digital Library
- [40] . 2015. Voxnet: A 3D convolutional neural network for real-time object recognition. In Proceedings of the international conference on Intelligent Robots and Systems (IROS’15). IEEE, 922–928.Google Scholar
Digital Library
- [41] . 2015. Exploiting silhouette descriptors and synthetic data for hand gesture recognition. In Proceedings of the Conference on Smart Tools & Apps for Graphics (STAG’15).Google Scholar
- [42] . 2003. Processing shape, motion and three-dimensional shape-from-motion in the human cortex. Cerebral Cortex 13, 5 (2003), 508–516.Google Scholar
Cross Ref
- [43] . 2012. Local phase quantization descriptor for improving shape retrieval/classification. Pattern Recogn. Lett. 33, 16 (2012), 2254–2260.Google Scholar
Digital Library
- [44] . 2017. Pointnet++: Deep hierarchical feature learning on point sets in a metric space. In Advances in Neural Information Processing Systems. MIT Press, 5099–5108.Google Scholar
Digital Library
- [45] . 2013. Robust part-based hand gesture recognition using Kinect sensor. IEEE Trans. Multimedia 15, 5 (
Aug. 2013), 1110–1120. Google ScholarDigital Library
- [46] . 2019. Surface diagnosticity predicts the high-level representation of regular and irregular object shape in human vision. Attent. Percept. Psychophys. 81 (2019), 1589–1608.Google Scholar
Cross Ref
- [47] . 2019. Design, large-scale usage testing, and important metrics for augmented reality gaming applications. ACM Trans. Multimedia Comput. Commun. Appl. 15, 2 (2019), 1–18.Google Scholar
Digital Library
- [48] . 2009. Fast point feature histograms (FPFH) for 3D registration. In Proceedings of the International Conference on Robotics and Automation. IEEE, 3212–3217.Google Scholar
Cross Ref
- [49] . 2008. Aligning point cloud views using persistent feature histograms. In Proceedings of the the International Conference on Intelligent Robots and Systems. IEEE, 3384–3391.Google Scholar
Cross Ref
- [50] . 2004. Recognition of shapes by editing their shock graphs. IEEE Trans. Pattern Anal. Mach. Intell. 26, 5 (2004), 550–571.Google Scholar
Digital Library
- [51] . 2001. Graph-based matching of occluded hand gestures. In Proceedings of the Applied Imagery Pattern Recognition Workshop. 67–73. Google Scholar
Cross Ref
- [52] . 1992. Feature-based correspondence: An eigenvector approach. Image Vision Comput. 10, 5 (1992), 283–288.Google Scholar
Digital Library
- [53] . 2013. The emerging field of signal processing on graphs: Extending high-dimensional data analysis to networks and other irregular domains. IEEE Signal Process. Mag. 30, 3 (
May 2013), 83–98. Google ScholarCross Ref
- [54] . 2008. Retrieving articulated 3D models using medial surfaces. Mach. Vision Appl. 19, 4 (2008), 261–275.Google Scholar
Digital Library
- [55] . 2016. Deep learning 3D shape surfaces using geometry images. In Proceedings of the European Conference on Computer Vision. Springer, 223–240.Google Scholar
Cross Ref
- [56] . 2015. Multi-view convolutional neural networks for 3D shape recognition. In Proceedings of the International Conference on Computer Vision. 945–953.Google Scholar
Digital Library
- [57] . 2011. Local shape descriptor selection for object recognition in range data. Comput. Vision Image Understand. 115, 5 (2011), 681–694.Google Scholar
Digital Library
- [58] . 2020. Convolutional neural network with spatial pyramid pooling for hand gesture recognition. Neural Comput. Appl. 33 (2021), 5339–5351.Google Scholar
- [59] . 1988. An eigendecomposition approach to weighted graph matching problems. IEEE Trans. Pattern Anal. Mach. Intell. 10, 5 (
Sept. 1988), 695–703. Google ScholarDigital Library
- [60] . 2002. Description of 3D-shape using a complex function on the sphere. In Proceedings of the International Conference on Multimedia and Expo, Vol. 1. IEEE, 177–180.Google Scholar
Cross Ref
- [61] . 2010. Shape classification using tree-unions. In Proceedings of the International Conference on of Pattern Recognition(ICPR’10). 983–986.Google Scholar
Digital Library
- [62] . 2015. Superpixel-based hand gesture recognition with Kinect depth camera. IEEE Trans. Multimedia 17, 1 (2015), 29–39.Google Scholar
Cross Ref
- [63] . 2020. Improving multiperson pose estimation by mask-aware deep reinforcement learning. ACM Trans. Multimedia Comput. Commun. Appl. 16, 3 (2020), 1–18.Google Scholar
Digital Library
- [64] . 1982. Contour curvature analysis: Hyperacuities in the discrimination of detailed shape. Vision Res. 22, 4 (1982), 449–460.Google Scholar
Cross Ref
- [65] . 2015. 3D shapenets: A deep representation for volumetric shapes. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR’15). 1912–1920.Google Scholar
- [66] . 2015. Deepshape: Deep learned shape descriptor for 3D shape matching and retrieval. In Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR’15). 1275–1283.Google Scholar
- [67] . 2016. Shape matching by part alignment using extended chordal axis transform. Pattern Recogn. 57 (2016), 115–135.Google Scholar
Digital Library
- [68] . 1972. Fourier descriptors for plane closed curves. IEEE Trans. Comput. 100, 3 (1972), 269–281.Google Scholar
Digital Library
- [69] . 2019. O2O method for fast 2D shape retrieval. IEEE Trans. Image Process. 28, 11 (2019), 5366–5378.Google Scholar
Digital Library
- [70] . 2018. Toward real-time 3D object recognition: A lightweight volumetric CNN framework using multitask learning. Comput. Graph. 71 (2018), 199–207.Google Scholar
Cross Ref
- [71] . 2015. Factorized graph matching. IEEE Trans. Pattern Anal. Mach. Intell. 38, 9 (2015), 1774–1789. Google Scholar
Digital Library
- [72] . 2017. 2D compressive sensing and multi-feature fusion for effective 3D shape retrieval. Info. Sci. 409 (2017), 101–120.Google Scholar
Cross Ref
Index Terms
GHOSM: Graph-based Hybrid Outline and Skeleton Modelling for Shape Recognition
Recommendations
Skeleton graph matching based on critical points using path similarity
ACCV'09: Proceedings of the 9th Asian conference on Computer Vision - Volume Part IIIThis paper proposes a novel graph matching algorithm based on skeletons and applies it to shape recognition based on object silhouettes The main idea is to match the critical points (junction points and end points) on skeleton graphs by comparing the ...
Shape recognition by bag of skeleton-associated contour parts
We associate a contour point with an object thickness value.We propose a new descriptor to combine both contour and skeleton information.Our method achieves the state-of-the-arts results on several shape dataset. Contour and skeleton are two ...
Skeleton-based Hierarchical Shape Segmentation
SMI '07: Proceedings of the IEEE International Conference on Shape Modeling and Applications 2007We present an effective framework for segmenting 3D shapes into meaningful components using the curve skeleton. Our algorithm identifies a number of critical points on the curve skeleton, either fully automatically as the junctions of the curve skeleton,...






Comments