Why Is the Census Transform Good for Robust Optic Flow Computation?

Hafner, David Zachary; Demetz, Oliver; Weickert, Joachim

doi:10.1007/978-3-642-38267-3_18

Cited by 59 publications

(30 citation statements)

References 22 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Nevertheless, this alternative census-inspired perspective offers an unexpected insight: As pointed out in [10], each binary digit of a census signature can be regarded as the sign of the corresponding directional derivative (in a finite difference sense). Thus, from this point of view, one can conclude that the complete rank transform inherently contains rich local differential information.…”

Section: Discussionmentioning

confidence: 99%

“…There are several recent publications that incorporate the census transform in variational optical flow or stereo methods: Müller et al propose a census-based data term for optical flow [15], and Ranftl et al [18] as well as Mei et al [12] present censusbased stereo methods. The theoretical study of Hafner et al [10] explains the reasons why census-based data terms for variational optical flow are successful.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

The Complete Rank Transform: A Tool for Accurate and Morphologically Invariant Matching of Structures

Demetz¹,

Hafner²,

Weickert³

2013

Procedings of the British Machine Vision Conference 2013

Self Cite

View full text Add to dashboard Cite

Most researchers agree that invariances are desirable in computer vision systems. However, one always has to keep in mind that this is at the expense of accuracy: By construction, all invariances inevitably discard information. The concept of morphological invariance is a good example for this trade-off and will be in the focus of this paper. Our goal is to develop a descriptor of local image structure that carries the maximally possible amount of local image information under this invariance. To fulfill this requirement, our descriptor has to encode the full ordering of the pixel intensities in the local neighbourhood. As a solution, we introduce the complete rank transform, which stores the intensity rank of every pixel in the local patch. As a proof of concept, we embed our novel descriptor in a prototypical TV−L 1 -type energy functional for optical flow computation, which we minimise with a traditional coarse-to-fine warping scheme. In this straightforward framework, we demonstrate that our descriptor is preferable over related features that exhibit the same invariance. Finally, we show by means of public benchmark systems that our method produces -in spite of its simplicity -results of competitive quality.

show abstract

Section: Discussionmentioning

confidence: 99%

Section: Related Workmentioning

confidence: 99%

The Complete Rank Transform: A Tool for Accurate and Morphologically Invariant Matching of Structures

Demetz¹,

Hafner²,

Weickert³

2013

Procedings of the British Machine Vision Conference 2013

Self Cite

View full text Add to dashboard Cite

show abstract

“…Census transform [283] recently regained interest and was promoted in [226] for optical flow estimation [108,118,179,182,183,205,252]. The Census signature is a bit string reflecting relative value of pixels of a patch with respect to the center pixel.…”

Section: Patch-based Similarity Measuresmentioning

confidence: 99%

“…Integrating the Census transform in variational optical flow is not trivial since it cannot be easily linearized. Solutions to remedy this problem are convex approximation [252], reformulation as a generalization of the gradient constancy conservation [108] or linearization of the data term [182,205] as previously mentioned for NCC [272].…”

Section: Patch-based Similarity Measuresmentioning

confidence: 99%

Optical flow modeling and computation: A survey

Fortun

Bouthémy

Kervrann

2015

Computer Vision and Image Understanding

338

182

View full text Add to dashboard Cite

a b s t r a c tOptical flow estimation is one of the oldest and still most active research domains in computer vision. In 35 years, many methodological concepts have been introduced and have progressively improved performances, while opening the way to new challenges. In the last decade, the growing interest in evaluation benchmarks has stimulated a great amount of work. In this paper, we propose a survey of optical flow estimation classifying the main principles elaborated during this evolution, with a particular concern given to recent developments. It is conceived as a tutorial organizing in a comprehensive framework current approaches and practices. We give insights on the motivations, interests and limitations of modeling and optimization techniques, and we highlight similarities between methods to allow for a clear understanding of their behavior.

show abstract

“…Considering the illumination variation on the KITTI dataset, we adopt Census Transform [37], which is proved to be powerful for robust optical flow computation [6]. We show the performance of different methods when integrated into cross-scale cost aggregation in Table 2.…”

Section: Kitti Datasetmentioning

confidence: 99%

Cross-Scale Cost Aggregation for Stereo Matching

Zhang

Fang

Min

et al. 2014

2014 IEEE Conference on Computer Vision and Pattern Recognition

115

View full text Add to dashboard Cite

Human beings process stereoscopic correspondence across multiple scales. However, this bio-inspiration is ignored by state-of-the-art cost aggregation methods for dense stereo correspondence. In this paper, a generic cross-scale cost aggregation framework is proposed to allow multi-scale interaction in cost aggregation. We firstly reformulate cost aggregation from a unified optimization perspective and show that different cost aggregation methods essentially differ in the choices of similarity kernels. Then, an inter-scale regularizer is introduced into optimization and solving this new optimization problem leads to the proposed framework. Since the regularization term is independent of the similarity kernel, various cost aggregation methods can be integrated into the proposed general framework. We show that the cross-scale framework is important as it effectively and efficiently expands state-of-the-art cost aggregation methods and leads to significant improvements, when evaluated on Middlebury, KITTI and New Tsukuba datasets.

show abstract

Why Is the Census Transform Good for Robust Optic Flow Computation?

Cited by 59 publications

References 22 publications

The Complete Rank Transform: A Tool for Accurate and Morphologically Invariant Matching of Structures

The Complete Rank Transform: A Tool for Accurate and Morphologically Invariant Matching of Structures

Optical flow modeling and computation: A survey

Cross-Scale Cost Aggregation for Stereo Matching

Contact Info

Product

Resources

About