Efficient 3D Object Recognition from Cluttered Point Cloud

Recognizing 3D objects and estimating their postures in a complex scene is a challenging task. Sample Consensus Initial Alignment (SAC-IA) is a commonly used point cloud-based method to achieve such a goal. However, its efficiency is low, and it cannot be applied in real-time applications. This paper analyzes the most time-consuming part of the SAC-IA algorithm: sample generation and evaluation. We propose two improvements to increase efficiency. In the initial aligning stage, instead of sampling the key points, the correspondence pairs between model and scene key points are generated in advance and chosen in each iteration, which reduces the redundant correspondence search operations; a geometric filter is proposed to prevent the invalid samples to the evaluation process, which is the most time-consuming operation because it requires transforming and calculating the distance between two point clouds. The introduction of the geometric filter can significantly increase the sample quality and reduce the required sample numbers. Experiments are performed on our own datasets captured by Kinect v2 Camera and on Bologna 1 dataset. The results show that the proposed method can significantly increase (10–30×) the efficiency of the original SAC-IA method without sacrificing accuracy.

Topical Issues in Publishing the Letters of F. M. Dostoevsky and His Correspondents

The article provides an overview of the editorial principles in publishing F.M. Dostoevsky’s epistolary heritage. An analysis of these principles allows to discuss the evolution of goals and objectives of the researchers involved in the preparation of Dostoevsky’s and his correspondents’ letters for publication. At the same time, this review reveals the problems that the researchers have encountered and continue to face for over 130 years of studies of the writer’s epistolary heritage. The literary value of Dostoevsky’s letters was recognized immediately after his death. They aroused interest both as a biographical source and as part of the writer’s literary heritage. A significant number of letters to A. E. Vrangel, A. N. Maykov, N. N. Strakhov, I. S. Aksakov and others were published as early as in the 1880s. Particular attention was heeded to Dostoevsky’s correspondence with readers, writers, editors, and journalists. The first publications of Dostoevsky’s letters did not pursue any scientific goals and presented them as part of the writer’s creative heritage. Many problems associated with the systematization of correspondence, search for manuscript autographs, lost or undiscovered letters, principles of publication of epistolary sources have already emerged at that time. In the 1920—1930s, the researchers, still focusing on Dostoevsky’s letters, turned to his addressees’ letters, which began to be recognized as an integral part of the correspondence. The corps of letters of certain correspondents were set apart, specific epistolary cycles were formed, mutual correspondence began to be published, and a gradual mastering of its historical, cultural and commentary potential commenced. The emerging trends were subsequently developed. Principles that included the completeness of presentation of correspondence, precision of reproduction of handwritten text, and a scientific and critical approach to the study and publication of the letters came to the fore. Despite the significant successes achieved by the researchers, many problems of publishing Dostoevsky’s correspondence are still relevant, for example, structuring the letters in an integral manner. At present, the publication of the writer’s epistolary heritage and the letters of his correspondents sets the task of publishing a complete annotated code of correspondence both in print and in an electronic form. The electronic publication format has certain advantages, as it expands the number of manuscript material presentation modes and, in turn, the chance for researchers to further study Dostoevsky’s epistolary heritage and the writer’s life and work as a whole.

SEMANTIC ROAD SCENE KNOWLEDGE FOR ROBUST SELF-CALIBRATION OF ENVIRONMENT-OBSERVING VEHICLE CAMERAS

<p><strong>Abstract.</strong> Environment-observing vehicle camera self-calibration using a structure from motion (SfM) algorithm allows calibration over vehicle lifetime without the need of special calibration objects being present in the calibration images. Scene-specific problems with feature-based correspondence search and reconstruction during the SfM pipeline might be caused by critical objects like moving objects, poor-texture objects or reflecting objects and might have negative influence on camera calibration. In this contribution, a method to use semantic road scene knowledge by means of semantic masks for a semantic-guided SfM algorithm is proposed to make the calibration more robust. Semantic masks are used to exclude image parts showing critical objects from feature extraction, whereby semantic knowledge is obtained by semantic segmentation of the road scene images. The proposed method is tested with an image sequence recorded in a suburban road scene. It has been shown that semantic guidance leads to smaller deviations of the estimated interior orientation and distortion parameters from reference values obtained by test field calibration compared to a standard SfM algorithm.</p>

BICOS—An Algorithm for Fast Real-Time Correspondence Search for Statistical Pattern Projection-Based Active Stereo Sensors

Pattern projection-based 3D measurement systems are widely used for contactless, non-destructive optical 3D shape measurements. In addition, many robot-operated automation tasks require real-time reconstruction of accurate 3D data. In previous works, we have demonstrated 3D scanning based on statistical pattern projection-aided stereo matching between two cameras. One major advantage of this technology is that the actually projected patterns do not have to be known a priori in the reconstruction software. This allows much simpler projector designs and enables high-speed projection. However, to find corresponding pixels between cameras, it is necessary to search the best match amongst all pixels within the geometrically possible image area (that is, within a range on the corresponding epipolar line). The well-established method for this search is to compare each candidate pixel by temporal normalized cross correlation of the brightness value sequences of both pixels. This is computationally expensive and interdicts fast real-time applications on inexpensive computer hardware. We show two variants of our algorithm “Binary Correspondence Search” (BICOS), which solve this task in significantly reduced calculation time. In practice, our algorithm is much faster than traditional, purely cross-correlation-based search while maintaining a similar level of accuracy.

Stereopsis is adaptive for the natural environment

Humans and many animals have forward-facing eyes providing different views of the environment. Precise depth estimates can be derived from the resulting binocular disparities, but determining which parts of the two retinal images correspond to one another is computationally challenging. To aid the computation, the visual system focuses the search on a small range of disparities. We asked whether the disparities encountered in the natural environment match that range. We did this by simultaneously measuring binocular eye position and three-dimensional scene geometry during natural tasks. The natural distribution of disparities is indeed matched to the smaller range of correspondence search. Furthermore, the distribution explains the perception of some ambiguous stereograms. Finally, disparity preferences of macaque cortical neurons are consistent with the natural distribution.