Ground Truth

With four decades of data on a population of bottlenose dolphins (Tursiops truncatus) resident to Sarasota Bay (SB), The Sarasota Dolphin Research Program offers an unparalleled platform for ground-truthing stable isotope data and exploring bottlenose dolphin ecology in a natural setting. We explored carbon isotope value fidelity to habitat utilization by comparing delta13C data from whole teeth and muscle to

Scene reconstruction from video sequences has become a prominent computer vision research area in recent years, due to its large number of applications in fields such as security, robotics and virtual reality. Despite recent progress in this field, there are still a number of issues that manifest as incomplete, incorrect or computationally-expensive reconstructions. The engine behind achieving reconstruction is the matching of features between images, where common conditions such as occlusions, lighting changes and texture-less regions can all affect matching accuracy. Subsequent processes that rely on matching accuracy, such as camera parameter estimation, structure computation and non-linear parameter optimization, are also vulnerable to additional sources of error, such as degeneracies and mathematical instability. Detection and correction of errors, along with robustness in parameter solvers, are a must in order to achieve a very accurate final scene reconstruction. However, err...

An investigation to localise facial landmarks from 3D images is presented, without using any assumption concerning facial pose. This paper introduces new surface descriptors, which are derived from either unstructured face data, or a radial basis function (RBF) model of the facial surface. Two new variants of feature descriptors are described, generally named as point - triplet descriptors because they require three vertices to be computed. The first is related to the classical depth map feature, which is referred to as weighted - interpolated depth map. The second variant of descriptors are derived from an implicit RBF model, they are referred to as surface RBF signature (SRS) features. Both variants of descriptors are able to encode surface information within a triangular region defined by a point - triplet into a surface signature, which could be useful not only for 3D face processing but also within a number of graph based retrieval applications. These descriptors are embedded into a system designed to localise the nose - tip and two inner - eye corners. Landmark localisation performance is reported by computing errors of estimated landmark locations against our respective ground --truth data from the Face Recognition Grand Challenge (FRGC) database.

Musical incipits are short extracts of scores, taken from the beginning. The RISM A/II collection contains about half a million of them. This large collection size makes a ground truth very interesting for the development of music retrieval methods, but at the same time makes it very difficult to establish one. Human experts cannot be expected to sift through half a million melodies to find the best matches for a given query. For 11 queries, we filtered the collection so that about 50 candidates per query were left, which we then presented to 35 human experts for a final ranking. We present our filtering methods, the experiment design, and the resulting ground truth. To obtain ground truths, we ordered the incipits by the median ranks assigned to them by the human experts. For every incipit, we used the Wilcoxon rank sum test to com-pare the list of ranks assigned to it with the lists of ranks assigned to its predecessors. As a result, we know which rank differences are statisticall...

This paper addresses the challenge of bridging the seman- tic gap between the rich meaning users desire when they query to locate and browse media and the shallowness of media descriptions that can be computed in today's content management systems. To facilitate high-level semantics-based content annotation and interpretation, we tackle the prob- lem of automatic decomposition of motion pictures into

Detection and classification of mine-like objects in side-scan sonar images needs to compensate for variability of objects, noise and background signatures. The unsupervised algorithm presented in this paper addresses improvements with respect to previous work and focuses on object and shadow detection based on morphological operators. Feature extraction from the detected objects and their classification into two classes, namely mine or non-mine like objects is described. Row-wise processing technique is applied for decreasing computational costs and memory usage to allow easy porting of the algorithm to an embedded architecture. The performance of the algorithms is measured against the obtained ground-truth.

A new empirical model for the retrieval, at a field scale, of the bare soil moisture content and the surface roughness characteristics from radar measurements is proposed. The derivation of the algorithm is based on the results of three experimental radar campaigns conducted under natural conditions over agricultural areas. Radar data were acquired by means of several C-band space borne

Shape matching is one of the main research procedures performed by archaeologists. Real time 3D computer graphics, 3D digitisation and content based retrieval are technologies that can facilitate the automation of the shape matching process. As pottery plays a significant role in understanding ancient societies, we focus on the development of compact shape descriptors that can be used for content based retrieval of complete or nearly complete 3D pottery replicas. In this work, we present shape descriptors that exploit the axial symmetry feature and attempt to enhance the archaeological study of pottery by using 3D graphics technologies. We evaluated the performance of the descriptors using a 3D pottery ground truth repository as a test bed. We created an experimental 3D pottery search engine and attempted to integrate a content based retrieval mechanism in a 3D virtual reality environment.

Hydrological monitoring is essential for meaningful water-management policies and actions, especially where water resources are scarce and/or dwindling, as is the case in Australia. In this paper, we investigate the regional 4° × 4° mascon (mass concentration) GRACE solutions for Australia provided by GSFC (Goddard Space Flight Center, NASA) for their suitability in monitoring Australian hydrology, with a particular focus on the Murray-Darling Basin (MDB). Using principal component analysis (PCA) and multi-linear regression analysis (MLRA), the main components of spatial and temporal variability in the mascon solutions are analysed over the whole Australian continent and the MDB. The results are compared to those from global solutions provided by CSR (Center for Space Research, University of Texas at Austin, USA) and CNES/GRGS (Centre National d'Études Spatiales/Groupe de Recherche de Geodesie Spatiale, France) and validated using data from the Tropical Rainfall Measuring Mission (TRMM), water storage changes predicted by the WaterGap Global Hydrological Model (WGHM) and the Global Land Data Assimilation System (GLDAS), and ground-truth (river-gauge) observations. For the challenging Australian case with generally weak hydrological signals, the mascon solutions provide similar results to those from the global solutions, with the advantage of not requiring additional filtering (destriping and smoothing) as, for example, is necessary for the CSR solutions. A further advantage of the mascon solutions is that they offer a higher temporal resolution (i.e., 10 days) compared to approximately monthly CSR solutions. Examining equivalent water volume (EWV) time series for the MDB shows a good cross-correlation (generally > 0.7) among the GRACE solutions when considering the whole basin, although lower (generally < 0.5) when all the GRACE solutions are compared to the TRMM, WGHM and GLDAS time series. Examining smaller portions of the MDB see the correlation among the GRACE solutions and the TRMM, WGHM and GLDAS EWV time series increase slightly (> 0.6), with all time series appearing to visually follow the general behaviour of the river-gauge data, although the cross-correlations are relatively low (between 0.3 and 0.6).► Mascon provides equivalent results as global CSR & CNES/GRGS solutions. ► All examined GRACE releases reveal seasonal & tropical north signals. ► GRACE, modelled hydrology & precipitation show similar behaviour Australia wide. ► GRACE solutions generally follow river gauge data.