Representing 3-D Objects in Range Images Using Geons

Quang-Loc, Nguyen; Levine, Martin D.

doi:10.1006/cviu.1996.0011

Cited by 16 publications

(4 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Others authors used the same representation, for example [17]. In the present work, bounding-boxes are adopted, which is a more abstract volumetric representation than commonly used parametrical models based on surfaces [18], [19], [20]. In addition, the quantization achieved by the voxel representation within the five-step time-span is related to occupancy grids [21], [22], but less computationally intensive.…”

Section: Related Workmentioning

confidence: 99%

Incremental object part detection toward object classification in a sequence of noisy range images

Gachter¹,

Harati²,

Siegwart³

2008

2008 IEEE International Conference on Robotics and Automation

View full text Add to dashboard Cite

Abstract-This paper presents an incremental object part detection algorithm using a particle filter. The method infers object parts from 3D data acquired with a range camera. The range information is uniquely quantized and enhanced by local structure information to partially cope with considerable measurement noise and distortion. The augmented voxel representation allows the adaptation of known track-before-detect algorithms to infer multiple object parts in a range image sequence. The appropriateness of the method is successfully demonstrated by an experiment.

show abstract

Section: Related Workmentioning

confidence: 99%

Incremental object part detection toward object classification in a sequence of noisy range images

Gachter¹,

Harati²,

Siegwart³

2008

2008 IEEE International Conference on Robotics and Automation

View full text Add to dashboard Cite

show abstract

“…Several types of primitives exist: generalized cylinders (Binford 1971;Ponce et al 1989;Zeroug and Nevatia 1996), geons (Biederman 1985;Nguyen and Levine 1996), superquadrics (Barr 1981), supershapes (Gielis et al 2003), hyperquadrics (Han et al 1993) and other implicit polynomial models (Keren et al 1994). These models can be classified in two main categories: quantitative models having a real representational power (superquadrics, hyperquadrics, implicit polynomials) and qualitative models which have a symbolic and structural representation purpose (geons, generalized cylinders).…”

Section: Primitive-based Modelsmentioning

confidence: 99%

Basic Background in 3D Object Processing

Lavoué¹

2008

3D Object Processing

View full text Add to dashboard Cite

This chapter details the different 3D representation models, commonly used in computer graphics and 3D modeling. We distinguish three main representation schemes: polygonal meshes; surface-based models, such as parametric, implicit and subdivision surfaces; and volumetric models including primitivebased models (superquadric, hyperquadric), voxel representation and constructive solid geometry. For each of these 3D representation schemes, we analyze the advantages and drawbacks in terms of modeling, regarding different applications.The point-based representation, which has been recently introduced in computer graphics, is not within the scope of this book. Recent reviews can be found in Kobbelt and Botsch (2004) and Alexa et al. (2004).

show abstract

“…However, in general \clean" or complete line drawings of objects cannot be obtained due to the colour and texture of object surfaces or complex illumination con gurations. Because of this, and also for practical reasons, some research has focussed on data obtained from laser range nders 31,25,24]. In all of the latter, part descriptions were determined in a bottom-up fashion, inferring global properties by aggregating local features.…”

Section: Part Identi Cationmentioning

confidence: 99%

“…This is illustrated in Figure 2 by ellipsoids. Transversality has been widely applied to part segmentation 30,31,11,34,35].…”

Section: Motivationmentioning

confidence: 99%

Signal-to-Symbol Mapping for Laser Rangefinders

Levine

1999

Handbook of Pattern Recognition and Computer Vision

Self Cite

View full text Add to dashboard Cite

A new approach for computing qualitative part-based descriptions of 3D objects is presented. The object descriptions are obtained in two steps: Object segmentation into parts and part model identi cation. Beginning with single-or multi-view range data of a 3D object, we simulate the charge density distribution over an object's surface which has been tessellated by a triangular mesh. We detect the deep surface concavities by tracing local charge density minima and then decompose the object into parts at these points. The individual parts are then modelled by parametric geons. The latter are seven qualitative shapes, each of which is formulated by a restricted globally deformed superellipsoid. Model recovery is performed by tting all parametric geons to a part and selecting the best model for the part, based on the minimum tting residual. A newly de ned objective function and a fast global optimisation technique are employed to obtain robust model tting results. Experiments demonstrate that this approach can successfully recover qualitative shape models from input data, especially when part shapes are not fully consistent with model shapes. The resultant object descriptions are well suited for symbolic reasoning and fast object recognition.

show abstract

Representing 3-D Objects in Range Images Using Geons

Cited by 16 publications

References 20 publications

Incremental object part detection toward object classification in a sequence of noisy range images

Incremental object part detection toward object classification in a sequence of noisy range images

Basic Background in 3D Object Processing

Signal-to-Symbol Mapping for Laser Rangefinders

Contact Info

Product

Resources

About