Multi-view stereo estimates the depth maps of multiple perspective images in a scene and then fuses them to generate 3D point cloud scene, which is an essential technology reconstruction. In this paper, we propose deep learning method GDINet, applying probabilistic methods pyramid framework, can significantly improve reconstruction quality. detail, first establish Gaussian distribution for each...

Age structure data is essential for single species stock assessments but length-frequency data can provide complementary information. In south-western Australia, the majority of these data for exploited species are derived from line caught fish. However, baited remote underwater stereo-video systems (stereo-BRUVS) surveys have also been found to provide accurate length measurements. Given that ...

W e address a n open and hitherto neglected problem i n computer vision, how to reconstruct the geometry of objects with arbitrary and possibly anisotropic bidirectional reflectance distribution functions (BRDFs) . Present reconstruction techniques, whether stereo vision, structure f rom motion, laser range finding, etc. make explicit or implicit assumptions about the B R D F . Here, we introdu...

Irradiation-induced anisotropic swelling in hexagonal α-SiC is known to degrade the mechanical properties of SiC; however, associated physical mechanism and microstructural process remain insufficiently understood. In this study, an condition where surface normal direction was allowed freely expand with constraint lateral introduced 4H-SiC using selected-area He+ irradiation, internal defect di...

A new approach for stereo reconstruction is proposed. This approach is based on a Gibbs probability distribution for surfaces in 3D space. The problem of stereo reconstruction is formulated then as a Bayes decision task. The main difference compared with known methods is the use of a more realistic cost function. In case of stereo reconstruction this function can be designed in some natural way...

This thesis mainly deals with large scale objects of height around 1.5 meters. It uses a dynamic stereo model or a binocular stereo model to calibrate and reconstruct a 3D large scale object. This thesis briefly introduces two basic calibration techniques, i.e. DLT calibration technique and Tsai’s calibration technique. It then discusses and evaluates the impact of different factors on the cali...

Parts or products from high performance metal are very expensive, partly due to the processing complexities during manufacturing. Recent studies have indicated that hybrid processes of additive manufacturing and machining process can be used to repair titanium parts, thus extending the service life. In order to implement these methods automatically, it is necessary to obtain the spatial geometr...

Non-destructive detection of wire bonding defects in integrated circuits (IC) is critical for ensuring product quality after packaging. Image-processing-based methods do not provide a detailed evaluation the three-dimensional wire. Therefore, method 3D reconstruction and pattern recognition based on stereo vision, which can achieve non-destructive proposed. The contour features wires other elec...

In this work we study a charged particle in the presence of both a continuous distribution of disclinations and a continuous distribution of edge dislocations in the framework of the geometrical theory of defects. We obtain the self-energy for a single charge both in the internal and external regions of either distribution. For both distributions the result outside the defect distribution is th...