/pdf/convex-analysisnoer-san-nojin-zhan-nituite-5dl2rbmoyr.pdf

Convex Analysisの二,三の進展について

We present an algorithm for fast global registration of partially overlapping 3D surfaces. The algorithm operates on candidate matches that cover the surfaces. A single objective is optimized to align the surfaces and disable false matches. The objective is defined densely over the surfaces and the optimization achieves tight alignment with no initialization. No correspondence updates or closest-point queries are performed in the inner loop. An extension of the algorithm can perform joint global registration of many partially overlapping surfaces. Extensive experiments demonstrate that the presented approach matches or exceeds the accuracy of state-of-the-art global registration pipelines, while being at least an order of magnitude faster. Remarkably, the presented approach is also faster than local refinement algorithms such as ICP. It provides the accuracy achieved by well-initialized local refinement algorithms, without requiring an initialization and at lower computational cost.

/pdf/fast-global-registration-1gxm3pt2yy.pdf

Fast Global Registration

We present an approach to indoor scene reconstruction from RGB-D video. The key idea is to combine geometric registration of scene fragments with robust global optimization based on line processes. Geometric registration is error-prone due to sensor noise, which leads to aliasing of geometric detail and inability to disambiguate different surfaces in the scene. The presented optimization approach disables erroneous geometric alignments even when they significantly outnumber correct ones. Experimental results demonstrate that the presented approach substantially increases the accuracy of reconstructed scene models.

/pdf/robust-reconstruction-of-indoor-scenes-1drcfrfeu2.pdf

Robust reconstruction of indoor scenes

We present a new simultaneous localization and mapping SLAM system capable of producing high-quality globally consistent surface reconstructions over hundreds of meters in real time with only a low-cost commodity RGB-D sensor. By using a fused volumetric surface reconstruction we achieve a much higher quality map over what would be achieved using raw RGB-D point clouds. In this paper we highlight three key techniques associated with applying a volumetric fusion-based mapping system to the SLAM problem in real time. First, the use of a GPU-based 3D cyclical buffer trick to efficiently extend dense every-frame volumetric fusion of depth maps to function over an unbounded spatial region. Second, overcoming camera pose estimation limitations in a wide variety of environments by combining both dense geometric and photometric camera pose constraints. Third, efficiently updating the dense map according to place recognition and subsequent loop closure constraints by the use of an 'as-rigid-as-possible' space deformation. We present results on a wide variety of aspects of the system and show through evaluation on de facto standard RGB-D benchmarks that our system performs strongly in terms of trajectory estimation, map quality and computational performance in comparison to other state-of-the-art systems.

/pdf/real-time-large-scale-dense-rgb-d-slam-with-volumetric-nsvs7jcopb.pdf

Real-time large-scale dense RGB-D SLAM with volumetric fusion

In recent years, great progress has been made on the problem of 3D scene reconstruction using depth sensors. On a large scale, these reconstructions look impressive, but often many fine details are lacking due to limitations in the sensor resolution. In this paper we combine two well-known principles for recovery of 3D models, namely fusion of depth images with photometric stereo to enhance the details of the reconstructions. We derive a simple and transparent objective functional that takes both the observed intensity images and depth information into account. The experimental results show that many details are captured that are not present in the input depth images. Moreover, we provide a quantitative evaluation that confirms the improvement of the resulting 3D reconstruction using a 3D printed model.

Combining Depth Fusion and Photometric Stereo for Fine-Detailed 3D Models

In many areas, the ability to create accurate 3D models is of great interest, for example, in computer vision, robotics, architecture, and augmented reality. In this paper we show how a textured indoor environment can be reconstructed in 3D using an RGB-D camera. Real-time performance can be achieved using a GPU. We show how the camera pose can be estimated directly using the geometry that we represent as a signed distance function (SDF). Since the SDF contains information about the distance to the surface, it defines an error-metric which is minimized to estimate the pose of the camera. By iteratively estimating the camera pose and integrating the new depth images into the model, the 3D reconstruction is computed on the fly. We present several examples of 3D reconstructions made from a handheld and robot-mounted depth sensor, including detailed reconstructions from medium-sized rooms with almost drift-free pose estimation. Furthermore, we demonstrate that our algorithm is robust enough for 3D reconstruction using data recorded from a quadrocopter, making it potentially useful for navigation applications. (Less)

/pdf/direct-camera-pose-tracking-and-mapping-with-signed-distance-6tlnhvq466.pdf

Direct Camera Pose Tracking and Mapping With Signed Distance Functions

One of the major research areas in computer vision is scene reconstruction from image streams. The advent of RGB-D cameras, such as the Microsoft Kinect, has lead to new possibilities for performing accurate and dense 3D reconstruction. There are already well-working algorithms to acquire 3D models from depth sensors, both for large and small scale scenes. However, these methods often break down when the scene geometry is not so informative, for example, in the case of planar surfaces. Similarly, standard image-based methods fail for texture-less scenes. We combine both color and depth measurements from an RGB-D sensor to simultaneously reconstruct both the camera motion and the scene geometry in a robust manner. Experiments on real data show that we can accurately reconstruct large-scale 3D scenes despite many planar surfaces.

Robust Camera Tracking by Combining Color and Depth Measurements

Online 3D reconstruction has been an active research area for a long time. Since the release of the Microsoft Kinect Camera and publication of KinectFusion [11] attention has been drawn how to acquire dense models in real-time. In this paper we present a method to make online 3D reconstruction which increases robustness for scenes with little structure information and little texture information. It is shown empirically that our proposed method also increases robustness when the distance between the camera positions becomes larger than what is commonly assumed. Quantitative and qualitative results suggest that this approach can handle situations where other well-known methods fail. This is important in, for example, robotics applications like when the camera position and the 3D model must be created online in real-time.

Robust online 3D reconstruction combining a depth sensor and sparse feature points

This paper tackles the problem of data abstraction in the context of 3D point sets. Our method classifies points into different geometric primitives, such as planes and cones, leading to a compact representation of the data. Being based on a semi-global Hough voting scheme, the method does not need initialization and is robust, accurate, and efficient. We use a local, low-dimensional parameterization of primitives to determine type, shape and pose of the object that a point belongs to. This makes our algorithm suitable to run on devices with low computational power, as often required in robotics applications. The evaluation shows that our method outperforms state-of-the-art methods both in terms of accuracy and robustness.

Erik Bylow

Papers

Combining Depth Fusion and Photometric Stereo for Fine-Detailed 3D Models

Direct Camera Pose Tracking and Mapping With Signed Distance Functions

Robust Camera Tracking by Combining Color and Depth Measurements

Robust online 3D reconstruction combining a depth sensor and sparse feature points

PrimiTect: Fast Continuous Hough Voting for Primitive Detection.