Results 1 - 10 of 24

What makes a chair a chair

by Helmut Grabner, Juergen Gall, Luc Van Gool - In CVPR , 2011
"... Many object classes are primarily defined by their functions. However, this fact has been left largely unexploited by visual object categorization or detection systems. We propose a method to learn an affordance detector. It identifies locations in the 3d space which “support ” the particular functi ..."
Abstract - Cited by 48 (1 self) - Add to MetaCart
Many object classes are primarily defined by their functions. However, this fact has been left largely unexploited by visual object categorization or detection systems. We propose a method to learn an affordance detector. It identifies locations in the 3d space which “support ” the particular function. Our novel approach “imagines ” an actor performing an action typical for the target object class, instead of relying purely on the visual object appearance. So, function is handled as a cue complementary to appearance, rather than being a consideration after appearance-based detection. Experimental results are given for the functional category “sitting”. Such affordance is tested on a 3d representation of the scene, as can be realistically obtained through SfM or depth cameras. In contrast to appearancebased object detectors, affordance detection requires only very few training examples and generalizes very well to other sittable objects like benches or sofas when trained on a few chairs. 1.
(Show Context)

Multiple view semantic segmentation for street view images

by Jianxiong Xiao, Long Quan - In ICCV
"... We propose a simple but powerful multi-view semantic segmentation framework for images captured by a camera mounted on a car driving along streets. In our approach, a pair-wise Markov Random Field (MRF) is laid out across multiple views. Both 2D and 3D features are extracted at a super-pixel level t ..."
Abstract - Cited by 43 (7 self) - Add to MetaCart
We propose a simple but powerful multi-view semantic segmentation framework for images captured by a camera mounted on a car driving along streets. In our approach, a pair-wise Markov Random Field (MRF) is laid out across multiple views. Both 2D and 3D features are extracted at a super-pixel level to train classifiers for the unary data terms of MRF. For smoothness terms, our approach makes use of color differences in the same image to identify accurate segmentation boundaries, and dense pixel-to-pixel correspondences to enforce consistency across different views. To speed up training and to improve the recognition quality, our approach adaptively selects the most similar training data for each scene from the label pool. Furthermore, we also propose a powerful approach within the same framework to enable large-scale labeling in both the 3D space and 2D images. We demonstrate our approach on more than 10,000 images from Google Maps Street View. 1.
(Show Context)

3-D Scene Analysis via Sequenced Predictions over Points and Regions

by Xuehan Xiong, Daniel Munoz, J. Andrew, Bagnell Martial Hebert
"... Abstract — We address the problem of understanding scenes from 3-D laser scans via per-point assignment of semantic labels. In order to mitigate the difficulties of using a graphical model for modeling the contextual relationships among the 3-D points, we instead propose a multi-stage inference proc ..."
Abstract - Cited by 32 (4 self) - Add to MetaCart
Abstract — We address the problem of understanding scenes from 3-D laser scans via per-point assignment of semantic labels. In order to mitigate the difficulties of using a graphical model for modeling the contextual relationships among the 3-D points, we instead propose a multi-stage inference procedure to capture these relationships. More specifically, we train this procedure to use point cloud statistics and learn relational information (e.g., tree-trunks are below vegetation) over fine (point-wise) and coarse (region-wise) scales. We evaluate our approach on three different datasets, that were obtained from different sensors, and demonstrate improved performance. I.
(Show Context)

Article Segmentation Based Classification of 3D Urban Point Clouds: A Super-Voxel Based Approach with Evaluation

by Ahmad Kamal Aijazi, Paul Checchin, Laurent Trassoudaine , 2013
"... Abstract: Segmentation and classification of urban range data into different object classes have several challenges due to certain properties of the data, such as density variation, inconsistencies due to missing data and the large data size that require heavy computation and large memory. A method ..."
Abstract - Cited by 6 (1 self) - Add to MetaCart
Abstract: Segmentation and classification of urban range data into different object classes have several challenges due to certain properties of the data, such as density variation, inconsistencies due to missing data and the large data size that require heavy computation and large memory. A method to classify urban scenes based on a super-voxel segmentation of sparse 3D data obtained from LiDAR sensors is presented. The 3D point cloud is first segmented into voxels, which are then characterized by several attributes transforming them into super-voxels. These are joined together by using a link-chain method rather than the usual region growing algorithm to create objects. These objects are then classified using geometrical models and local descriptors. In order to evaluate the results, a new metric that combines both segmentation and classification results simultaneously is presented. The effects of voxel size and incorporation of RGB color and laser reflectance intensity on the classification results are also discussed. The method is evaluated on standard data sets using different metrics to demonstrate its efficacy. Keywords: 3D objects
(Show Context)

Learning to Hash Logistic Regression for Fast 3D Scan Point Classification

by Jens Behley, Kristian Kersting, Dirk Schulz, Volker Steinhage, Armin B. Cremers
"... Abstract — Segmenting range data into semantic categories has become a more and more active field of research in robotics. In this paper, we advocate to view this task as a problem of fast, large-scale retrieval. Intuitively, given a dataset of millions of labeled scan points and their neighborhoods ..."
Abstract - Cited by 3 (0 self) - Add to MetaCart
Abstract — Segmenting range data into semantic categories has become a more and more active field of research in robotics. In this paper, we advocate to view this task as a problem of fast, large-scale retrieval. Intuitively, given a dataset of millions of labeled scan points and their neighborhoods, we simply search for similar points in the datasets and use the labels of the retrieved ones to predict the labels of a novel point using some local prediction model such as majority vote or logistic regression. However, actually carrying this out requires highly efficient ways of (1) storing millions of scan points in memory and (2) quickly finding similar scan points to a target scan point. In this paper, we propose to address both issues by employing Weiss et al.’s recent spectral hashing. It represents each item in a database by a compact binary code that is constructed so that similar items will have similar binary code words. In turn, similar neighbors have codes within a small Hamming distance of the code for the query. Then, we learn a logistic regression model locally over all points with the same binary code word. Our experiments on real world 3D scans show that the resulting approach, called spectrally hashed logistic regression, can be ultra fast at prediction time and outperforms state-of-the art approaches such as logistic regression and nearest neighbor. I.
(Show Context)

Vehicle Detection from Low Quality Aerial LIDAR Data

by Bo Yang, Pramod Sharma, Ram Nevatia
"... In this paper we propose a vehicle detection framework on low resolution aerial range data. Our system consists of three steps: data mapping, 2D vehicle detection and postprocessing. First, we map the range data into 2D grayscale images by using the depth information only. For this purpose we propos ..."
Abstract - Cited by 2 (0 self) - Add to MetaCart
In this paper we propose a vehicle detection framework on low resolution aerial range data. Our system consists of three steps: data mapping, 2D vehicle detection and postprocessing. First, we map the range data into 2D grayscale images by using the depth information only. For this purpose we propose a novel local ground plane estimation method, and the estimated ground plane is further refined by a global refinement process. Then we compute the depth value of missing points (points for which no depth information is available) by an effective interpolation method. In the second step, to train a classifier for the vehicles, we describe a method to generate more training examples from very few training annotations and adopt the fast cascade Adaboost approach for detecting vehicles in 2D grayscale images. Finally, in post-processing step we design a novel method to detect some vehicles which are comprised of clusters of missing points. We evaluate our method on real aerial data and the experiments demonstrate the effectiveness of our approach. Figure 1. An example Aerial Data taken from [12]. 1.
(Show Context)

Label Propagation from ImageNet to 3D Point Clouds

by Yan Wang, Rongrong Ji, Shih-fu Chang
"... Recent years have witnessed a growing interest in understanding the semantics of point clouds in a wide variety of applications. However, point cloud labeling remains an open problem, due to the difficulty in acquiring sufficient 3D point labels towards training effective classifiers. In this paper, ..."
Abstract - Cited by 2 (1 self) - Add to MetaCart
Recent years have witnessed a growing interest in understanding the semantics of point clouds in a wide variety of applications. However, point cloud labeling remains an open problem, due to the difficulty in acquiring sufficient 3D point labels towards training effective classifiers. In this paper, we overcome this challenge by utilizing the existing massive 2D semantic labeled datasets from decadelong community efforts, such as ImageNet and LabelMe, and a novel “cross-domain ” label propagation approach. Our proposed method consists of two major novel components, Exemplar SVM based label propagation, which effectively addresses the cross-domain issue, and a graphical model based contextual refinement incorporating 3D constraints. Most importantly, the entire process does not require any training data from the target scenes, also with good scalability towards large scale applications. We evaluate our approach on the well-known Cornell Point Cloud Dataset, achieving much greater efficiency and comparable accuracy even without any 3D training data. Our approach shows further major gains in accuracy when the training data from the target scenes is used, outperforming state-ofthe-art approaches with far better efficiency. 1.
(Show Context)

Characterization of 3-D volumetric probabilistic scenes for object recognition

by Maria I. Restrepo, On A. Mayer, Student Member, Ali O. Ulusoy, Graduate Student Member, Joseph L. Mundy - IEEE J , 2012
"... Abstract—This paper presents a new volumetric representation for categorizing objects in large-scale 3-D scenes reconstructed from image sequences. This work uses a probabilistic volumetric model (PVM) that combines the ideas of background modeling and volumetric multi-view reconstruction to handle ..."
Abstract - Cited by 2 (0 self) - Add to MetaCart
Abstract—This paper presents a new volumetric representation for categorizing objects in large-scale 3-D scenes reconstructed from image sequences. This work uses a probabilistic volumetric model (PVM) that combines the ideas of background modeling and volumetric multi-view reconstruction to handle the uncertainty inherent in the problem of reconstructing 3-D structures from 2-D images. The advantages of probabilistic modeling have been demonstrated by recent application of the PVM representation to video image registration, change detection and classification of changes based on PVM context. The applications just mentioned, operate on 2-D projections of the PVM. This paper presents the first work to characterize and use the local 3-D information in the scenes. Two approaches to local feature description are proposed and compared: 1) features derived from a PCA analysis of model neighborhoods; and 2) features derived from the coefficients of a 3-D Taylor series expansion within each neighborhood. The resulting description is used in a bag-of-features approach to classify buildings, houses, cars, planes, and parking lots learned from aerial imagery collected over Providence, RI. It is shown that both feature descriptions explain the data with similar ac-curacy and their effectiveness for dense-feature categorization is compared for the different classes. Finally, 3-D extensions of the Harris corner detector and a Hessian-based detector are used to detect salient features. Both types of salient features are evaluated through object categorization experiments, where only features with maximal response are retained. For most saliency criteria tested, features based on the determinant of the Hessian achieved higher classification accuracy than Harris-based features. Index Terms—3-D data processing, 3-D object recognition, ma-chine vision, Bayesian learning. I.
(Show Context)

3D Shape Registration

by Umberto Castellani, Adrien Bartoli
"... Registration is the problem of bringing together two or more 3D shapes, either of the same object or of two different but similar objects. This chapter first introduces the classical Iterative Closest Point (ICP) algorithm which represents the gold standard registration method. Current limitations o ..."
Abstract - Cited by 1 (0 self) - Add to MetaCart
Registration is the problem of bringing together two or more 3D shapes, either of the same object or of two different but similar objects. This chapter first introduces the classical Iterative Closest Point (ICP) algorithm which represents the gold standard registration method. Current limitations of ICP are addressed and the most popular variants of ICP are described to improve the basic implementation in several ways. Challenging registration scenarios are analyzed and a taxonomy of recent and promising alternative registration techniques is introduced. Three case studies are then described with an increasing level of difficulty. The first case study describes a simple but effective technique to detect outliers. The second case study uses the Levenberg-Marquardt optimization procedure to solve standard pairwise registration. The third case study focuses on the challenging problem of deformable object registration. Finally, open issues and directions for future work are discussed and conclusions are drawn. 1
(Show Context)

Segmentation and matching: Towards a robust object detection system

by Jing Huang, Suya You - In Winter Conference on Applications of Computer Vision (WACV
"... This paper focuses on detecting parts in laser-scanned data of a cluttered industrial scene. To achieve the goal, we propose a robust object detection system based on seg-mentation and matching, as well as an adaptive segmenta-tion algorithm and an efficient pose extraction algorithm based on corres ..."
Abstract - Cited by 1 (1 self) - Add to MetaCart
This paper focuses on detecting parts in laser-scanned data of a cluttered industrial scene. To achieve the goal, we propose a robust object detection system based on seg-mentation and matching, as well as an adaptive segmenta-tion algorithm and an efficient pose extraction algorithm based on correspondence filtering. We also propose an overlapping-based criterion that exploits more information of the original point cloud than the number-of-matching cri-terion that only considers key-points. Experiments show how each component works and the results demonstrate the performance of our system compared to the state of the art. 1.
(Show Context)