View variation of point-set and line-segment features

Burns, J. Brian; Weiss, Richard; Riseman, Edward M.

doi:10.1109/34.184774

Cited by 128 publications

(56 citation statements)

References 15 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…That is, unless we could eliminate the nuisance ν without "throwing away information" on the scene ξ. 5 Unfortunately, Corollary 1 says that this is not possible: in order to achieve viewpoint invariance, shape information has to be sacrificed. In light of this result, then, does it still make sense to use features?…”

Section: Why Features?mentioning

confidence: 99%

See 1 more Smart Citation

Features for recognition: viewpoint invariance for non-planar scenes

Vedaldi

Soatto

2005

Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1

View full text Add to dashboard Cite

We present a technique for local image representation that is invariant to viewpoint for scenes with arbitrary non-planar shape. We show that generic viewpoint invariance can be achieved, under suitable conditions, although the resulting invariant is not shape-discriminative. Our results serve both to validate existing approaches to local feature detection and description, as well as to complement them where they are not applicable. We illustrate our approach on images of 3-D corners where existing approaches fail.

show abstract

Section: Why Features?mentioning

confidence: 99%

“…On viewpoint invariance, Burns et al [5] showed that there do not exist generic viewpoint invariants. This statement, however, is misleading, since it refers to collections of points in space, with no photometric signature associated to them.…”

Section: State Of the Art And Our Contributionsmentioning

confidence: 99%

Features for recognition: viewpoint invariance for non-planar scenes

Vedaldi

Soatto

2005

Tenth IEEE International Conference on Computer Vision (ICCV'05) Volume 1

View full text Add to dashboard Cite

show abstract

“…We first present a proof from Clemens and .Jacobs):32] which applies only to scaled orthographic projection, and then a proof discovered by Burns, Weiss and Riseman [24] and by Moses and Ullman [84] which applies to all projection models. The results of Clemens and Jacobs [31] and Burns, \Weiss and Riseman [22] appeared simultaneously. The work of Moses and Ulhlan [83] appeared later.…”

Section: There Are No Invariantsmentioning

confidence: 99%

Recognizing 3-D Objects Using 2-D Images

Jacobs¹

1993

View full text Add to dashboard Cite

“…It is well known [Burns et al 1990] that this cannot be done without some external, or "modelbased", assumption, namely prior information about the shape of the object. This is easy to see if we have a 1-D view of a 2-D point set.…”

Section: -D Shapesmentioning

confidence: 99%

Geometric Invariants and Object Recognition.

Weiss¹

1992

View full text Add to dashboard Cite

IIn this paper we discuss the role of the general invariance concept in object recognition, and review the classical and recent literature on projective invariance. Invariants help solve major problems of object recognition. For instance, different images of the same object often differ from each other, because of the different viewpoint from which they were taken. To match the two images, common methods thus need to find the correct viewpoint, a difficult problem that can involve search in a large parameter space of all possible points of view and/or finding point correspondences. Geometric invariants are shape descriptors, computed from the geometry of the shape, that remain unchanged under geometric transformations such as changing the viewpoint. Thus they can be matched without search. Deformations of objects are another important class of changes for which invariance is useful.

show abstract

View variation of point-set and line-segment features

Cited by 128 publications

References 15 publications

Features for recognition: viewpoint invariance for non-planar scenes

Features for recognition: viewpoint invariance for non-planar scenes

Recognizing 3-D Objects Using 2-D Images

Geometric Invariants and Object Recognition.

Contact Info

Product

Resources

About