This paper presents a new objectclassification technique for 3d point clouddata acquired with a laser scanner. In general, it is not straightforward to distinguish objects that have similar 3d structures but belong ...
详细信息
ISBN:
(数字)9781510611221
ISBN:
(纸本)9781510611214;9781510611221
This paper presents a new objectclassification technique for 3d point clouddata acquired with a laser scanner. In general, it is not straightforward to distinguish objects that have similar 3d structures but belong to different categories based only on the range data. To tackle this issue, we focus on laser reflectance obtained as a side product of range measurement by a laser scanner. Since laser reflectance contains appearance information, the proposed method classifies objects based on not only geometrical features in range data but also appearance features in reflectance data, both of which are acquired by a single laser scanner. Furthermore, we extend the conventional Histogram of Oriented Gradients (HOG) so that it couples geometrical and appearance information more tightly. Experiments show the proposed technique combining geometrical and appearance information outperforms conventional techniques.
In this paper we describe our 3dobject signature for 3d object classification. The signature is based on a learning approach that finds salient points on a 3dobject and represent these points in a 2d spatial map bas...
详细信息
ISBN:
(纸本)9783540896883
In this paper we describe our 3dobject signature for 3d object classification. The signature is based on a learning approach that finds salient points on a 3dobject and represent these points in a 2d spatial map based on a longitude-latitude transformation. Experimental results show high classification rates on both pose-normalized and rotatedobjects and include a study on classification accuracy as a function of number of rotations in the training set.
This paper proposes a volumetric part based3d object classification approach. Superquadric-hased Geon (SBG) description is implemented for representing individual volumetric parts, the constituents of 3dobject. The ...
详细信息
ISBN:
(纸本)9781424404759
This paper proposes a volumetric part based3d object classification approach. Superquadric-hased Geon (SBG) description is implemented for representing individual volumetric parts, the constituents of 3dobject. The classification of 3d ohject is decomposed into the constrained search on interpretation tree and the similarity measure computation. A set of integrated features and corresponding constraints are presented, which not only reflect individual parts' shape, but model's topological information among volumetric parts. These constraints are used to direct an efficient tree search. Following the searching stage, a similarity measure computation algorithm is developed to evaluate the shape similarity of objectdata and the stored models. By this classification approach, both whole and partial matching results with similarity ranks can he obtained;especially, focus match can be achieved, in which different key parts can be labeled and all the matched models with corresponding key parts can be obtained Some experiments are given to show the validity and efficiency of the approach for 3d object classification.
This paper proposes a volumetric part based3d object classification approach. Superquadric-based Geon (SBG) description is implemented for representing individual volumetric parts, the constituents of 3dobject. The ...
详细信息
ISBN:
(纸本)9781424404759
This paper proposes a volumetric part based3d object classification approach. Superquadric-based Geon (SBG) description is implemented for representing individual volumetric parts, the constituents of 3dobject. The classification of 3dobject is decomposed into the constrained search on interpretation tree and the similarity measure computation. A set of integrated features and corresponding constraints are presented, which not only reflect individual parts' shape, but model's topological information among volumetric parts. These constraints are used to direct an efficient tree search. Following the searching stage, a similarity measure computation algorithm is developed to evaluate the shape similarity of objectdata and the stored models. By this classification approach, both whole and partial matching results with similarity ranks can be obtained;especially, focus match can be achieved, in which different key parts can be labeled and all the matched models with corresponding key parts can be obtained. Some experiments are given to show the validity and efficiency of the approach for 3d object classification.
暂无评论