Terms Of Surface Reconstruction Research Articles

Implicit Conversion of Manifold B-Rep Solids by Neural Halfspace Representation

We present a novel implicit representation --- neural halfspace representation (NH-Rep), to convert manifold B-Rep solids to implicit representations. NH-Rep is a Boolean tree built on a set of implicit functions represented by the neural network, and the composite Boolean function is capable of representing solid geometry while preserving sharp features. We propose an efficient algorithm to extract the Boolean tree from a manifold B-Rep solid and devise a neural network-based optimization approach to compute the implicit functions. We demonstrate the high quality offered by our conversion algorithm on ten thousand manifold B-Rep CAD models that contain various curved patches including NURBS, and the superiority of our learning approach over other representative implicit conversion algorithms in terms of surface reconstruction, sharp feature preservation, signed distance field approximation, and robustness to various surface geometry, as well as a set of applications supported by NH-Rep.

Object-RPE: Dense 3D reconstruction and pose estimation with convolutional neural networks

We present an approach for recognizing objects present in a scene and estimating their full pose by means of an accurate 3D instance-aware semantic reconstruction. Our framework couples convolutional neural networks (CNNs) and a state-of-the-art dense Simultaneous Localization and Mapping (SLAM) system, ElasticFusion (Whelan et al., 2016), to achieve both high-quality semantic reconstruction as well as robust 6D pose estimation for relevant objects. We leverage the pipeline of ElasticFusion as a backbone, and propose a joint geometric and photometric error function with per-pixel adaptive weights. While the main trend in CNN-based 6D pose estimation has been to infer object’s position and orientation from single views of the scene, our approach explores performing pose estimation from multiple viewpoints, under the conjecture that combining multiple predictions can improve the robustness of an object detection system. The resulting system is capable of producing high-quality instance-aware semantic reconstructions of room-sized environments, as well as accurately detecting objects and their 6D poses. The developed method has been verified through extensive experiments on different datasets. Experimental results confirmed that the proposed system achieves improvements over state-of-the-art methods in terms of surface reconstruction and object pose prediction. Our code and video are available at https://sites.google.com/view/object-rpe.

Panoptic 3D Mapping and Object Pose Estimation Using Adaptively Weighted Semantic Information

We present a system capable of reconstructing highly detailed object-level models and estimating the 6D pose of objects by means of an RGB-D camera. In this work, we integrate deep-learning-based semantic segmentation, instance segmentation, and 6D object pose estimation into a state of the art RGB-D mapping system. We leverage the pipeline of ElasticFusion as a backbone and propose modifications of the registration cost function to make full use of the semantic class labels in the process. The proposed objective function features tunable weights for the depth, appearance, and semantic information channels, which are learned from data. A fast semantic segmentation and registration weight prediction convolutional neural network (Fast-RGBD-SSWP) suited to efficient computation is introduced. In addition, our approach explores performing 6D object pose estimation from multiple viewpoints supported by the high-quality reconstruction system. The developed method has been verified through experimental validation on the YCB-Video dataset and a dataset of warehouse objects. Our results confirm that the proposed system performs favorably in terms of surface reconstruction, segmentation quality, and accurate object pose estimation in comparison to other state-of-the-art systems. Our code and video are available at https://sites.google.com/view/panoptic-mope .

The Application Of Open-Source And Free Photogrammetric Software For The Purposes Of Cultural Heritage Documentation

Abstract The documentation of cultural heritage is an essential part of appropriate care of historical monuments, representing a part of our history. At present, it represents the current issue, for which considerable funds are being spent, as well as for the documentation of immovable historical monuments in a form of castle ruins, among the others. Non-contact surveying technologies - terrestrial laser scanning and digital photogrammetry belong to the most commonly used technologies, by which suitable documentation can be obtained, however their use may be very costly. In recent years, various types of software products and web services based on the SfM (or MVS) method and developed as open-source software, or as a freely available and free service, relying on the basic principles of photogrammetry and computer vision, have started to get into the spotlight. By using the services and software, acquired digital images of a given object can be processed into a point cloud, serving directly as a final output or as a basis for further processing. The aim of this paper, based on images of various objects of the Slanec castle ruins obtained by the DSLR Pentax K5, is to assess the suitability of different types of open-source and free software and free web services and their reliability in terms of surface reconstruction and photo-texture quality for the purposes of castle ruins documentation.

Review Article: Mechanism of Resonant Enhancement of Gold- and Copper-Nanowires Arrays

Although second harmonic generations from gold- and copper-nanowires have been the subject of extensive studies, a complete understanding of the mechanism of the enhancement is still missing. The aim of review article is to address these issues by performing systemic studies of mechanism of the resonant enhancements between gold- and copper-nanowires in terms of surface reconstruction and relativistic effect, and implements these types of candidate for the potential applications in optical communications and optical devices.

Catalytic Reactions on Crystallites

FIM and PFDMS are used for studying particles that mimic catalytic grains. Exposing tips to reactive gas mixtures allows following in-situ catalytic reactions on the nanometer scale. These techniques appear as unique tools to approach simultaneously the behavior of each plane of small crystallites in terms of surface reconstruction, adsorption and catalytic activity. They are complementary to environmental transmission and scanning electron microscopies.

Anomalous Auger electron spectroscopy results of chemical vapor deposited diamond (111) and (100) surfaces

High resolution Auger electron spectroscopy was performed on (111) and (100) facets as well as on a microvoid of polycrystalline diamond films. Auger fine structure analyses of these facets indicate anomalous line shapes when electron beam was focused to about 0.2 μm diameter in contrast to the large area scans (beam size ∼10×8 μm) which exhibit a typical Auger pattern characteristic of diamond. This unusual behavior is discussed in terms of surface reconstruction which may occur due to Auger assisted desorption of hydrogen.

Stability of (114) and (114) facets in III–V compounds under usual MBE conditions

Tensile GaInAs layers grown by MBE on InP(001) with lattice mismatch of −0.5% to −1.7% have been examined by transmission electron mictoscopy (TEM). Elastic relaxation occurs by drastic thickness modulations which are built first on (114) and (114) facets, then on (113) and (113) facets. Many examples of spontaneous development of (114) facets from (001) surface have also been reported by other groups. This is observed under usual MBE conditions and for various chemical compositions and lattice mismatches, positive as well as negative, for instance in compressively strained (Ga)InAs islands grown on GaAs(001) or InP(001). This is discussed in terms of surface reconstruction. It is suggested that (114) and (114) facets could be reconstructed in a way very similar to the usual 2 × 4 reconstruction of the (001) GaAsAs surface which is considered very stable. This is due to their crystallographic structure which involves exactly two gallium steps per 2 × 4 unit cell.

Influence of stress and surface reconstruction on the morphology of tensile GaInAs grown on InP(001) by gas source molecular beam epitaxy

Tensile/compressive and tensile/lattice matched Ga 1 − x In x As multilayers have been grown by gas source molecular beam epitaxy (GS-MBE) on InP(001) and examined by transmission electron microscopy (TEM). The tensile layers exhibit undulations parallel to [1 1 0] for a lattice mismatch of −0.5% and prismatic mesas for lattice mismatches of −1 and −1.7%. The facets making up the mesas are mainly of the (114)A and (113)A types. These growth modes are accounted for by elastic relaxation mechanisms. The relationship between the various morphologies and lattice mismatch is discussed by reviewing the competition between elastic energy and surface energy. It is shown that the relaxation can be partially or fully frozen by decreasing the growth temperature or increasing the V/III flux ratio. The dependence upon the growth technique is also underlined by comparison with metalorganic vapor phase epitaxy (MOVPE). The modulation anisotropy and the frequent occurrence of (114)A facets are examined in terms of surface reconstruction. It is suggested that the crystallographic morphology of (114)A facets facilitates a reconstruction which is very similar to the regular 2 × 4 reconstruction of the (001) GaAsAs surface. Therefore the exact nature of steps and facets and their surface reconstruction should be taken into account in elastic relaxation mechanisms.

Structure of Ga0.47In0.53As epitaxial layers grown on InP substrates at different temperatures

The evolution of defect structures and surface reconstruction of Ga0.47In0.53As epitaxial layers grown by molecular beam epitaxy on InP substrate have been investigated by TEM and RHEED over a wide growth temperature range (150 °C≤Tg≤450 °C) before and after annealing. In the growth temperature range 400 °C≤Tg≤450 °C, extensive segregation of In near the layer surface takes place. The maximum of In concentration was found to lie under the layer surface. The kinetic of surface pit formation after annealing was explained in terms of surface reconstruction and As clustering. Extensive As clustering was observed after annealing for the growth temperature range 175 °C≤Tg≤300 °C. Analysis of diffuse scattering and dark-field images made it possible to propose an atomic model of cluster structure. Precipitate formation close to the interface was observed only for the samples grown at 150 °C and annealed at 500 °C. Differences in electrical properties between LT-grown GaAs and Ga0.47In0.53As are explained in terms of the structural defects.