Search

This paper presents an approach for 3D geometry-based object categorization in cluttered table-top scenes. In our method, objects are decomposed into different geometric parts whose spatial arrangement is represented by a graph. The matching and searching of graphs representing the objects is sped up by using a hash table which contains possible spatial configurations of the different parts that constitute the objects. Additive feature descriptors are used to label partially or completely visible object parts. In this work we categorize objects into five geometric shapes: sphere, box, flat, cylindrical, and disk/plate, as these shapes represent the majority of objects found on tables in typical households. Moreover, we reconstruct complete 3D models that include the invisible back-sides of objects as well, in order to facilitate manipulation by domestic service robots. Finally, we present an extensive set of experiments on point clouds of objects using an RGBD camera, and our results highlight the improvements over previous methods. © 2014 Springer Science+Business Media Dordrecht.

Contour detection is an important characteristic of human vision perception. Humans can easily find the objects contour in a complex visual scene; however, traditional computer vision cannot do well. This paper primarily concerned with how to track the objects contour using a human-like vision. In this article, we propose a biologically motivated computational model to track and detect the objects contour. Even the previous research has proposed some models by using the Dijkstra algorithm 1, our work is to mimic the human eye movement and imitate saccades in our humans. We use natural images with associated ground truth contour maps to assess the performance of the proposed operator regarding the detection of contours while suppressing texture edges. The results show that our method enhances contour detection in cluttered visual scenes more effectively than classical edge detectors proposed by other methods. © Springer-Verlag Berlin Heidelberg 2012.

In this paper a feasibility study of liver CT dataset classification, using features from different scales of the wavelet transform analysis in conjunction with statistical pattern recognition methods is presented. In our study 850 extracted sub-images from 19 liver CT scans were used. Statistical measurements were collected; from the sub-images as well as from their different scale wavelet transform coefficients. We found by using the Leave-One-Out method that the combination of the features from the first and second Order statistics, achieved overall classification accuracy >90.0, both specificity and sensitivity >90.0. Features selected by the spatial domain performed better than the wavelet based techniques, under the classification rule of Quadratic Classifier (QC). In addition, features selected by the third scale wavelet transform coefficients performed better than those collected from the other wavelet scales, under the classification rule of Bayesian Classifier (BC).

The Adelson-Bergen motion energy sensor is well established as the leading model of low-level visual motion sensing in human vision. However, the standard model cannot predict adaptation effects in motion perception. A previous paper Pavan et al.(Journal of Vision 10:1-17, 2013) presented an extension to the model which uses a first-order RC gain-control circuit (leaky integrator) to implement adaptation effects which can span many seconds, and showed that the extended model's output is consistent with psychophysical data on the classic motion after-effect. Recent psychophysical research has reported adaptation over much shorter time periods, spanning just a few hundred milliseconds. The present paper further extends the sensor model to implement rapid adaptation, by adding a second-order RC circuit which causes the sensor to require a finite amount of time to react to a sudden change in stimulation. The output of the new sensor accounts accurately for psychophysical data on rapid forms of facilitation (rapid visual motion priming, rVMP) and suppression (rapid motion after-effect, rMAE). Changes in natural scene content occur over multiple time scales, and multi-stage leaky integrators of the kind proposed here offer a computational scheme for modelling adaptation over multiple time scales. © 2014 Springer Science+Business Media New York.

Multi-view range image integration focuses on producing a single reasonable 3D point cloud from multiple 2.5D range images for the reconstruction of a watertight manifold surface. However, registration errors and scanning noise usually lead to a poor integration and, as a result, the reconstructed surface cannot have topology and geometry consistent with the data source. This paper proposes a novel method cast in the framework of Markov random fields (MRF) to address the problem. We define a probabilistic description of a MRF labeling based on all input range images and then employ loopy belief propagation to solve this MRF, leading to a globally optimised integration with accurate local details. Experiments show the advantages and superiority of our MRF-based approach over existing methods.

A combined magnetic resonance (MR) imaging method has been proposed to investigate individual's knee functionality quantitatively under weight-bearing condition. High resolution MR data were acquired first to reconstruct the subject-specific anatomical model. A dynamic MR acquisition was obtained afterwards to record the motion of knee joint. A tri-rigid registration was applied to retrieve the knee joint motion, leading to a 12 degree-of-freedom (DoF) knee functional model. Using this model, the tibiofemoral contact mechanism was studied and analysed in both 2D and 3D. A mathematical definition of contact points of cartilage surfaces is given by modelling these surfaces as manifolds. It is believed that such subject-specific motion of contact points on cartilage surfaces of femur and two tibia plateaus can provide valuable insights for clinical applications such as knee replacement surgery.