Direct estimation of structure from nonlinear motion by voting algorithm without tracking and matching

Hamano, Teruo; Yasuno, Takato; Ishii, Kenichiro

doi:10.1109/icpr.1992.201610

Cited by 4 publications

(5 citation statements)

References 5 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Thresholding peak counts requires the position of the occupied voxel to be consistent when evidence is interpreted from different viewpoints. This second accumulation yields more accurate reconstructions [3]. It follows similar HT extensions for line detection [13].…”

Section: A Accumulation Based Modeling (Abm)mentioning

confidence: 59%

“…HT-domain interference between nearby features has also been noted in standard applications and in HT-based stereo [3] [4]. Reducing aliasing in the HT representation can improve the results of separate anti-interference techniques.…”

Section: Introductionmentioning

confidence: 98%

“…These methods often avoid correspondence matching by interpreting data in an object-centric 3D model. Some methods, like ABM, extend the Hough transform (HT) to fuse data in a voxel representation [3] [4]. A survey of multi-view techniques is given in [5] and the HT in [6].…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Multi-view stereo ranging via Distributed Ray Tracing

Coffman

Bovik

2010

2010 IEEE Southwest Symposium on Image Analysis &Amp; Interpretation (SSIAI)

View full text Add to dashboard Cite

We explore the use of Distributed Ray Tracing (DRT), an anti-aliasing technique from computer graphics, in multi-view computational stereo. As an example, we study ABM, a multi-view stereo algorithm based on a set of Hough transform accumulation operations. Augmenting ABM with DRT improves both internal signal quality and reconstruction accuracy. Results are given for both fundamental and complex "super-resolution reconstruction" tasks, where the voxel side length is less than the image ground sample distance. DRT improves ABM accuracy by 18% and can be generalized to improve other stereo algorithms.

show abstract

Section: A Accumulation Based Modeling (Abm)mentioning

confidence: 59%

Section: Introductionmentioning

confidence: 98%

See 1 more Smart Citation

Multi-view stereo ranging via Distributed Ray Tracing

Coffman

Bovik

2010

2010 IEEE Southwest Symposium on Image Analysis &Amp; Interpretation (SSIAI)

View full text Add to dashboard Cite

show abstract

“…Let us compare the cost of computation in the proposed method and the method proposed by Hamano et al [13], assuming that the Scene space is divided initially into do x do x do. After each voting, each voxel is divided into 8 subvoxels.…”

Section: Cost Of Computationmentioning

confidence: 99%

Multiscale structure from multiviews by ▽²G filtered three‐dimensional voting

Mori

Suzuki

Horikoshi

et al. 1995

Systems & Computers in Japan

View full text Add to dashboard Cite

SUMMARYA hierarchical matching is the most efficient way to recognize a 3-D object. In most conventional methods of 3-D object recognition, a hierarchical structure is formed after the detailed 3D structure of an object has been extracted by dividing it coarsely. This approach is very inefficient, since it is difficult to obtain a detailed 3D structure initially in addition to a large amount of computation time. This paper proposes a method which extracts a multiscale 3D structure directly from coarse to fine. In this method, a space containing an object is divided into coarse voxels, and a "3D voting" is applied. This voting adds a certain value to a voxel through which a back-projected line connects the center of the camera lens and a feature point on an image. A voxel having a high value is regarded as a region containing a 3D feature point, and this is extracted. Ektraction of a hierarchical structure is carried out by repeating the dividing process from coarse voxels to fine voxels so that the operation becomes efficient. This paper also analyzes the effects of the deviation of a feature point in an image and the deviation of the camera center on the 3D voting. The paper shows also that an application of a v2G filter to the voxel space (after 3D voting) is effective in suppressing the errors due to the deviations. The method has been confirmed successfully by processing six complex test images.

show abstract

“…The octree generation method needs image segmentation to obtain silhouette images [2]. Compared to these methods, Hamano's method which divides the scene space into voxels and votes for them with sight lines of the camera has fewer limitations [5].…”

Section: Introductionmentioning

confidence: 99%

Multi-scale structure from multi-views by Δ/sup 2/G filtered 3D voting

Mori

Suzuki

Horikoshi

et al.

Proceedings of IEEE Conference on Computer Vision and Pattern Recognition

Self Cite

View full text Add to dashboard Cite

into voxels. The values of all voxels are set to aero. This paper proposes a multi-scale 9D structure recovery method which divides the scene space into vozels and votes for them using the sight lines of the camem from coarse to fine. This paper also analyzes the mechanism behind the infEuence of the image quantization error and the position error of the camera's optical center. Both errors can be diminished by v'G filtering the voted wozel space. A multi-scale 3D structure of objects is successfully generated from a real amage sequence of 6 frames using this method.

show abstract

Direct estimation of structure from nonlinear motion by voting algorithm without tracking and matching

Cited by 4 publications

References 5 publications

Multi-view stereo ranging via Distributed Ray Tracing

Multi-view stereo ranging via Distributed Ray Tracing

Multiscale structure from multiviews by ▽²G filtered three‐dimensional voting

Multi-scale structure from multi-views by Δ/sup 2/G filtered 3D voting

Contact Info

Product

Resources

About

Direct estimation of structure from nonlinear motion by voting algorithm without tracking and matching

Cited by 4 publications

References 5 publications

Multi-view stereo ranging via Distributed Ray Tracing

Multi-view stereo ranging via Distributed Ray Tracing

Multiscale structure from multiviews by ▽2G filtered three‐dimensional voting

Multi-scale structure from multi-views by Δ/sup 2/G filtered 3D voting

Contact Info

Product

Resources

About

Multiscale structure from multiviews by ▽²G filtered three‐dimensional voting