Diffusion Geometry Derived Keypoints and Local Descriptors for 3D Deformable Shape Analysis

Author(s):  
Xupeng Wang ◽  
Mohammed Bennamoun ◽  
Ferdous Sohel ◽  
Hang Lei

Geometric analysis of three-dimensional (3D) surfaces with local deformations is a challenging task, required by mobile devices. In this paper, we propose a new local feature-based method derived from diffusion geometry, including a keypoint detector named persistence-based Heat Kernel Signature (pHKS), and a feature descriptor named Heat Propagation Strips (HeaPS). The pHKS detector first constructs a scalar field using the heat kernel signature function. The scalar field is generated at a small scale to capture fine geometric information of the local surface. Persistent homology is then computed to extract all the local maxima from the scalar field, and to provide a measure of persistence. Points with a high persistence are selected as pHKS keypoints. In order to describe a keypoint, an intrinsic support region is generated by the diffusion area. This support region is more robust than its geodesic distance counterpart, and provides a local surface with adaptive scale for subsequent feature description. The HeaPS descriptor is then developed by encoding the information contained in both the spatial and temporal domains of the heat kernel. We conducted several experiments to evaluate the effectiveness of the proposed method. On the TOSCA Dataset, the HeaPS descriptor achieved a high performance in terms of descriptiveness. The feature detector and descriptor were then tested on the SHREC 2010 Feature Detection and Description Dataset, and produced results that were better than the state-of-the-art methods. Finally, their application to shape retrieval was evaluated. The proposed pHKS detector and HeaPS descriptor achieved a notable improvement on the SHREC 2014 Human Dataset.

Author(s):  
Hui Zeng ◽  
Haipeng Yu ◽  
Wenli Liu ◽  
Wei Sun

This paper presents an effective 3D local feature descriptor, which is called Gradient Direction Accumulation-based Heat Kernel Signature (GDA-HKS) descriptor, and its application in nonrigid 3D model retrieval. The GDA-HKS descriptor is based on the heat kernel signature, and it is scale invariant and robust to the nonrigid deformation of the 3D model. Compared with the SI-HKS descriptor, the GDA-HKS descriptor is constructed directly in the time domain, and it can effectively avoid the loss of high frequency information. The absolute gradient difference is used to encode the GDA-HKS descriptor, which can describe the changing trend of the one-dimensional signal more effectively. Extensive experimental results have validated the effectiveness of the designed GDA-HKS descriptor.


2018 ◽  
Vol 2018 ◽  
pp. 1-13
Author(s):  
Huanyu Yang ◽  
Kuangrong Hao ◽  
Yongsheng Ding

A novel approach of 3D human model segmentation is proposed, which is based on heat kernel signature and geodesic distance. Through calculating the heat kernel signature of the point clouds of human body model, the local maxima of thermal energy distribution of the model is found, and the set of feature points of the model is obtained. Heat kernel signature has affine invariability which can be used to extract the correct feature points of the human model in different postures. We adopt the method of geodesic distance to realize the hierarchical segmentation of human model after obtaining the semantic feature points of human model. The experimental results show that the method can overcome the defect of geodesic distance feature extraction. The human body models with different postures can be obtained with the model segmentation results of human semantic characteristics.


2011 ◽  
Vol 41 (11) ◽  
pp. 2155-2167 ◽  
Author(s):  
Xavier Sanchez ◽  
Elena Roget ◽  
Jesus Planella ◽  
Francesc Forcat

Abstract The theoretical models of Batchelor and Kraichnan, which account for the smallest scales of a scalar field passively advected by a turbulent fluid (Prandtl > 1), have been validated using shear and temperature profiles measured with a microstructure profiler in a lake. The value of the rate of dissipation of turbulent kinetic energy ɛ has been computed by fitting the shear spectra to the Panchev and Kesich theoretical model and the one-dimensional spectra of the temperature gradient, once ɛ is known, to the Batchelor and Kraichnan models and from it determining the value of the turbulent parameter q. The goodness of the fit between the spectra corresponding to these models and the measured data shows a very clear dependence on the degree of isotropy, which is estimated by the Cox number. The Kraichnan model adjusts better to the measured data than the Batchelor model, and the values of the turbulent parameter that better fit the experimental data are qB = 4.4 ± 0.8 and qK = 7.9 ± 2.5 for Batchelor and Kraichnan, respectively, when Cox ≥ 50. Once the turbulent parameter is fixed, a comparison of the value of ɛ determined from fitting the thermal gradient spectra to the value obtained after fitting the shear spectra shows that the Kraichnan model gives a very good estimate of the dissipation, which the Batchelor model underestimates.


2013 ◽  
Vol 6 (3) ◽  
pp. 527-537 ◽  
Author(s):  
E. Jäkel ◽  
M. Wendisch ◽  
B. Mayer

Abstract. Spectral airborne upward and downward irradiance measurements are used to derive the area-averaged surface albedo. Real surfaces are not homogeneous in their reflectivity. Therefore, this work studies the effects of the heterogeneity of surface reflectivity on the area-averaged surface albedo to quantify how well aircraft measurements can resolve the small-scale variability of the local surface albedo. For that purpose spatially heterogeneous surface albedo maps were input into a 3-dimensional (3-D) Monte Carlo radiative transfer model to simulate 3-D irradiance fields. The calculated up- and downward irradiances in altitudes between 0.1 and 5 km are used to derive the area-averaged surface albedo using an iterative retrieval method that removes the effects due to atmospheric scattering and absorption within the layer beneath the considered level. For the case of adjacent land and sea surfaces, parametrizations are presented which quantify the horizontal distance from the coastline that is required to reduce surface heterogeneity effects on the area-averaged surface albedo to a given limit. The parametrization which is a function of altitude, aerosol optical depth, single scattering albedo, and the ratio of local land and sea albedo was applied for airborne spectral measurements. In addition, the deviation between area-averaged and local surface albedo is determined for more complex surface albedo maps. For moderate aerosol conditions (optical depth less than 0.4) and a wavelength range between 400 and 1000 nm, the altitude and the heterogeneity of the surface albedo are the dominant factors determining the mean deviation between local and area-averaged surface albedo. A parametrization of the mean deviation is applied to an albedo map that was derived from a Landsat image of an area in East Anglia (UK). Parametrization and direct comparison of local and area-averaged surface albedo show similar mean deviations (20% vs. 25%) over land.


Entropy ◽  
2020 ◽  
Vol 22 (11) ◽  
pp. 1290
Author(s):  
Hongjuan Gao ◽  
Guohua Geng ◽  
Sheng Zeng

Computer-aided classification serves as the basis of virtual cultural relic management and display. The majority of the existing cultural relic classification methods require labelling of the samples of the dataset; however, in practical applications, there is often a lack of category labels of samples or an uneven distribution of samples of different categories. To solve this problem, we propose a 3D cultural relic classification method based on a low dimensional descriptor and unsupervised learning. First, the scale-invariant heat kernel signature (Si-HKS) was computed. The heat kernel signature denotes the heat flow of any two vertices across a 3D shape and the heat diffusion propagation is governed by the heat equation. Secondly, the Bag-of-Words (BoW) mechanism was utilized to transform the Si-HKS descriptor into a low-dimensional feature tensor, named a SiHKS-BoW descriptor that is related to entropy. Finally, we applied an unsupervised learning algorithm, called MKDSIF-FCM, to conduct the classification task. A dataset consisting of 3D models from 41 Tang tri-color Hu terracotta Eures was utilized to validate the effectiveness of the proposed method. A series of experiments demonstrated that the SiHKS-BoW descriptor along with the MKDSIF-FCM algorithm showed the best classification accuracy, up to 99.41%, which is a solution for an actual case with the absence of category labels and an uneven distribution of different categories of data. The present work promotes the application of virtual reality in digital projects and enriches the content of digital archaeology.


2018 ◽  
Vol 483 (1) ◽  
pp. 289-298 ◽  
Author(s):  
Victor H Robles ◽  
James S Bullock ◽  
Michael Boylan-Kolchin
Keyword(s):  

2013 ◽  
Vol 23 (5) ◽  
pp. 1505-1522 ◽  
Author(s):  
Suk-Hwan Lee ◽  
Ki-Ryong Kwon ◽  
Won-Joo Hwang ◽  
V. Chandrasekar

2009 ◽  
Vol 630 ◽  
pp. 225-265 ◽  
Author(s):  
ISAAC W. EKOTO ◽  
RODNEY D. W. BOWERSOX ◽  
THOMAS BEUTNER ◽  
LARRY GOSS

The response of the mean and turbulent flow structure of a supersonic high-Reynolds-number turbulent boundary layer flow subjected to local and global mechanical distortions was experimentally examined. Local disturbances were introduced via small-scale wall patterns, and global distortions were induced through streamline curvature-driven pressure gradients. Local surface topologies included k-type diamond and d-type square elements; a smooth wall was examined for comparison purposes. Three global distortions were studied with each of the three surface topologies. Measurements included planar contours of the mean and fluctuating velocity via particle image velocimetry, Pitot pressure profiles, pressure sensitive paint and Schlieren photography. The velocity data were acquired with sufficient resolution to characterize the mean and turbulent flow structure and to examine interactions between the local surface roughness distortions and the imposed pressure gradients on the turbulence production. A strong response to both the local and global distortions was observed with the diamond elements, where the effect of the elements extended into the outer regions of the boundary layer. It was shown that the primary cause for the observed response was the result of local shock and expansion waves modifying the turbulence structure and production. By contrast, the square elements showed a less pronounced response to local flow distortions as the waves were significantly weaker. However, the frictional losses were higher for the blunter square roughness elements. Detailed quantitative characterizations of the turbulence flow structure and the associated production mechanisms are described herein. These experiments demonstrate fundamental differences between supersonic and subsonic rough-wall flows, and the new understanding of the underlying mechanisms provides a scientific basis to systematically modify the mean and turbulence flow structure all the way across supersonic boundary layers.


2013 ◽  
Vol 333-335 ◽  
pp. 969-973
Author(s):  
Yu Han Yang ◽  
Yao Qin Xie

To improve the efficiency and accuracy of the conventional SIFT-TPS (Scale-invariant feature transform and Thin-Plate Spline) method in deformable registration for CT lung image, we develop a novel approach by using combining SURF(Speeded up Robust Features) and GDLOH(Gradient distance-location-orientation histogram) to detect matching feature points. First, we employ SURF as feature detection to find the stable feature points of the two CT images rapidly. Then GDLOH is taken as feature descriptor to describe each detected points characteristic, in order to supply measurement tool for matching process. In our experiment, five couples of clinical images are simulated using our algorithm above, result in an obvious improvement in run-time and registration quality, compared with the conventional methods. It is demonstrated that the proposed method may create a new window in performing a good robust and adaptively for deformable registration for CT lung tomography.


Sign in / Sign up

Export Citation Format

Share Document