In this paper, we study the problem of 3d object segmentation from raw point clouds. Unlike existing methods which usually require a large amount of human annotations for full supervision, we propose the first unsuper...
详细信息
In this paper, we study the problem of 3d object segmentation from raw point clouds. Unlike existing methods which usually require a large amount of human annotations for full supervision, we propose the first unsupervised method, called OGC, to simultaneously identify multiple 3dobjects in a single forward pass, without needing any type of human annotations. The key to our approach is to fully leverage the dynamic motion patterns over sequential point clouds as supervision signals to automatically discover rigidobjects. Our method consists of three major components, 1) the objectsegmentation network to directly estimate multi-object masks from a single point cloud frame, 2) the auxiliary self-supervised scene flow estimator, and3) our core object geometry consistency component. By carefully designing a series of loss functions, we effectively take into account the multi-object rigid consistency and the object shape invariance in both temporal and spatial scales. This allows our method to truly discover the object geometry even in the absence of annotations. We extensively evaluate our method on five datasets, demonstrating the superior performance for object part instance segmentation and general objectsegmentation in both indoor and the challenging outdoor scenarios.
This paper aims to develop an automatic 3d object segmentation method for the large-scale point clouds. Given a range image, the preprocessing is first applied to get the optimal 3d point cloud. A k-nearest neighbor i...
详细信息
This paper aims to develop an automatic 3d object segmentation method for the large-scale point clouds. Given a range image, the preprocessing is first applied to get the optimal 3d point cloud. A k-nearest neighbor is built, and a segmentation algorithm based on the conditional angular clustering technique is used to segment the objects from the point cloud. The algorithm is tested on the real point clouddatasets. The experiment results demonstrated that the developedsegmentation method can be used to localize the object with the relative uncertainty of 0.27%.
3d reconstruction is difficult to use in general applications because it treats objects as a whole without distinction. Post-processing to segment them into individual objects is essential. However, when the reconstru...
详细信息
Numerous depth image-based rendering algorithms have been proposed to synthesize the virtual view for the free viewpoint television. However, inaccuracies in the depth map cause visual artifacts in the virtual view. I...
详细信息
Numerous depth image-based rendering algorithms have been proposed to synthesize the virtual view for the free viewpoint television. However, inaccuracies in the depth map cause visual artifacts in the virtual view. In this paper, we propose a novel virtual view synthesis framework to create the virtual view of the scene. Here, we incorporate a trilateral depth filter with local texture information, spatial proximity, and color similarity to remove the ghost contours by rectifying the misalignment between the depth map and its associated color image. To further enhance the quality of the synthesized virtual views, we partition the scene into different 3dobject segments based on the color image anddepth map. Each 3dobject segment is warped and blended independently to avoid mixing the pixels belonging to different parts of the scene. The evaluation results indicate that the proposed method significantly improves the quality of the synthesized virtual view compared with other methods and are qualitatively very similar to the ground truth. In addition, it also performs well in real-world scenes.
The point cloudsegmentation of a substation device attached with cables is the basis of substation identification and reconstruction. However, it is limited by a number of factors including the huge amount of point c...
详细信息
The point cloudsegmentation of a substation device attached with cables is the basis of substation identification and reconstruction. However, it is limited by a number of factors including the huge amount of point clouddata of a substation device, irregular shape, unclear feature distinction due to the auxiliary point clouddata attached to the main body of a device. Therefore, the segmentation efficiency of a substation device is very low. In order to improve the accuracy and efficiency of the point cloudsegmentation, this paper proposes a method to segment the attached cables point cloud of a substation device by using the shape feature of point cloud. Firstly, according to the spatial position of the point cloud of a substation device, octree is used to conduct voxelization of the point cloud, and the point cloud resampling is operated according to point clouddensity of each voxel, so as to reduce original point clouddata and improve computing efficiency. Then Mean Shift algorithm is used to locate the center axis of the point cloud, and cylinder growth method is used to initially segment cables data and locate the end of each cable. Finally, points of the end are used as seed points to carry out a region growth based on shape feature of the point cloud to realize effective segmentation of cables data. In the experiment, 303 sets of point cloud of devices are selected, including circuit breaker, voltage transformer, transformer, etc. The final result shows that the successful segmentation rate of this method reaches 95.34%, which effectively proves the feasibility of this method.
With the increasing degree of automation of indoor mobile robots, the perception requirements of indoor objects are also higher and higher, which needs to clarify the types of objects and master their three-dimensiona...
详细信息
ISBN:
(纸本)9798350386783;9798350386776
With the increasing degree of automation of indoor mobile robots, the perception requirements of indoor objects are also higher and higher, which needs to clarify the types of objects and master their three-dimensional size information. In this scheme, the combination of lidar and monocular cameras is adopted to compensate for the deficiency of a single sensor. The data-cascading network structure is adopted. Firstly, the YOLOP model detects and segments the image in two dimensions. After that, the PointNet model performs instance segmentation on the object's point cloud, followed by 3d bounding box regression to identify the target object. The network is used to determine the types of objects and obtain their three-dimensional size information. Through data set and field verification, the proposeddata cascade network is superior to most previous methods. It can accomplish the detection task of indoor 3dobjects based on lidar and monocular cameras.
3dobject recognition is a challenging task for intelligent and robot systems in industrial and home indoor *** is critical for such systems to recognize and segment the 3dobject instances that they encounter on a fr...
详细信息
3dobject recognition is a challenging task for intelligent and robot systems in industrial and home indoor *** is critical for such systems to recognize and segment the 3dobject instances that they encounter on a frequent *** computer vision,graphics,and machine learning fields have all given it a lot of ***,3dsegmentation was done with hand-crafted features anddesigned approaches that didn’t achieve acceptable performance and couldn’t be generalized to large-scale *** learning approaches have lately become the preferred method for 3dsegmentation challenges by their great success in 2d computer ***,the task of instance segmentation is currently less *** this paper,we propose a novel approach for efficient 3d instance segmentation using red green blue anddepth(RGB-d)data based on deep *** 2d region based convolutional neural networks(Mask R-CNN)deep learning model with point based rending module is adapted to integrate with depth information to recognize and segment 3d instances of *** order to generate 3d point cloud coordinates(x,y,z),segmented 2d pixels(u,v)of recognizedobject regions in the RGB image are merged into(u,v)points of the depth ***,we conducted an experiment and analysis to compare our proposed method from various points of view and *** experimentation shows the proposed3dobject recognition and instance segmentation are sufficiently beneficial to support object handling in robotic and intelligent systems.
In this work, an ACO-based approach to the problem of 3d object segmentation is presented. Ant Colony Optimization (ACO) meta-heuristic uses a set of agents to explore a search space, gathering local information and u...
详细信息
ISBN:
(数字)9783642386794
ISBN:
(纸本)9783642386787
In this work, an ACO-based approach to the problem of 3d object segmentation is presented. Ant Colony Optimization (ACO) meta-heuristic uses a set of agents to explore a search space, gathering local information and utilizing their common memory to obtain global solutions. In our approach to the 3dsegmentation problem, the artificial ants start their exploratory movements in the outer contour of the object. They explore the surface of the object influenced by its curvature and by the trails followed by other agents. After a number of generations, particular solutions of the agents converge to the best global paths, which are used as borders to segment the object's parts. This convergence mechanism avoids over-segmentation, detecting regions based on the global structure of the object and not on local information only.
due to the advantages of 3d point clouds over 2d optical images, the related researches on scene understanding in 3d point clouds have been increasingly attracting wide attention from academy and industry. However, ma...
详细信息
due to the advantages of 3d point clouds over 2d optical images, the related researches on scene understanding in 3d point clouds have been increasingly attracting wide attention from academy and industry. However, many 3d scene understanding methods largely require abundant supervised information for training a data-driven model. The acquisition of such supervised information relies on manual annotations which are laborious and arduous. Therefore, to mitigate such manual efforts for annotating training samples, this paper studies a unified neural network to segment 3dobjects out of point clouds interactively. Particularly, to improve the segmentation performance on the accurate objectsegmentation, the boundary information of 3dobjects in point clouds are encoded as a boundary energy term in the Markov Random Field (MRF) model. Moreover, the MRF model with the boundary energy term is naturally integrated with the Graphical Neural Network (GNN) to obtain a compact representation for generating the boundary-preserved3dobjects. The proposed method is evaluated on two point clouds datasets obtained from different types of laser scanning systems, i.e. terrestrial laser scanning system and mobile laser scanning system. Comparative experiments show that the proposed method is superior and effective in 3dobjects segmentation in different point-cloud scenarios.
This paper presents a novel objectsegmentation approach for highly complex indoor scenes. Our approach starts with a novel algorithm which partitions the scene into distinct regions whose boundaries accurately confor...
详细信息
This paper presents a novel objectsegmentation approach for highly complex indoor scenes. Our approach starts with a novel algorithm which partitions the scene into distinct regions whose boundaries accurately conform to the physical object boundaries in the scene. Next, we propose a novel perceptual grouping algorithm based on local cues (e.g., 3d proximity, co-planarity, and shape convexity) to merge these regions into object hypotheses. Our extensive experimental evaluations demonstrate that our objectsegmentation results are superior compared to the state-of-the-art methods.
暂无评论