Content-Based Retrieval of 3-D Objects Using Spin Image Signatures

Assfalg, J.; Bertini, Marco; Bimbo, Alberto Del; Pala, Pietro

doi:10.1109/tmm.2006.886271

Cited by 81 publications

(52 citation statements)

References 23 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…From Fig. 9a it is also worth noting that the performance of the proposed approach when using shape information only is rather similar to that of [17], consistently with the fact that both methods are based on the same type of descriptor. Instead the performance of the proposed method is clearly better than that of [17] when both shape and color information are combined together.…”

Section: Resultssupporting

confidence: 61%

“…9a it is also worth noting that the performance of the proposed approach when using shape information only is rather similar to that of [17], consistently with the fact that both methods are based on the same type of descriptor. Instead the performance of the proposed method is clearly better than that of [17] when both shape and color information are combined together. The performance difference between the two methods is a good indication of the improvement color information may bring to 3D retrieval (the method of [17] exploits shape data only as most methods in the literature).…”

Section: Resultssupporting

confidence: 58%

“…The proposed scheme outperforms the other methods in retrieving all the other gazebos, that indeed have a similar color structure, while its performance is not particularly impressive in retrieving the other buildings, which have a rather different color structure than those of the gazebos. Comparison of our approach, [22] and [17] over the query models set. a) Mean interpolated-precision vs. recall trends.…”

Section: Resultsmentioning

confidence: 99%

“…9a shows the precision vs. recall average on a set of queries on the PSB. The plot compares the proposed method against those for which we had an available implementation, i.e., the light field descriptor based method of [22] and the spin image retrieval scheme of [17] (the results of this method have been kindly provided by the authors). The normalized discounted cumulated gain trends for the same set of queries are shown in Fig.…”

Section: Resultsmentioning

confidence: 99%

“…These methods exploit descriptors representing the surface in the proximity of a vertex, like the bag of features based approach of [16], the 3D Shape Spectrum Descriptor used within the MPEG-7 framework or the spin-image descriptor, firstly introduced in [3] for object matching and then used for 3D retrieval in the schemes of [17] and [18]. There are also methods that combine both local and global descriptors, e.g., in [19] the global 3D shape is represented by the spatial configuration of a set of specific local shapes.…”

Section: Related Workmentioning

confidence: 99%

See 4 more Smart Citations

Combining color and shape descriptors for 3D model retrieval

Pasqualotto

Zanuttigh

Cortelazzo

2013

Signal Processing: Image Communication

View full text Add to dashboard Cite

Nowadays many three dimensional models feature color information together with the shape description. However current content-based retrieval schemes for 3D models are based on shape information only and ignore color clues. The significance of shape versus color clues for 3D model retrieval is instead a fundamental issue still almost unexplored at this time. A possible approach is to extend shape-based 3D model retrieval methods of proven effectiveness in order to include color. This work follows such rationale and introduces an extended version of the spin-image descriptor that can account also for color data. The comparison of color descriptors is performed using a novel scheme that allows to recognize as similar also objects with different colors but distributed in the same way over the shape. Shape and color similarity are finally combined together by an algorithm based on fuzzy logic. Experimental results show how the joint use of color and shape data allows to obtain better results than each of the two types of information alone. Comparisons with state-of-the-art content-based retrieval methods for 3D models also show how the proposed scheme outperforms standard solutions on object classes with meaningful color information.

show abstract

Section: Resultssupporting

confidence: 61%