A robust and efficient motion segmentation based on orthogonal projection matrix of shape space

Ichimura, Naoyuki

doi:10.1109/cvpr.2000.854877

Cited by 11 publications

(15 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The algorithm they suggested (as well as those suggested in [3,5,1,8]) relied on the block diagonal structure of Q which occurs if and only if V is block diagonal. However, the columns of V are…”

Section: Figure 1 Degeneracies In Multi-body (Single Sequence) Factomentioning

confidence: 99%

“…Hence, V and therefore Q will have a block diagonal structure if and only if the motion matrices M 1 and M 2 are linearly independent. When M 1 and [3,5,1,8], which rely on the block diagonal structure of Q will fail to separate between the objects. Note, that partial dependence occurs even if only a single column of the motion matrix of one object is linearly dependent on the columns of the motion matrix of the other object.…”

Section: Figure 1 Degeneracies In Multi-body (Single Sequence) Factomentioning

confidence: 99%

“…The body of work on multi-body factorization (e.g., [3,4,5,1,8]), suggest employing multi-frame linear subspace constraints to separate between objects moving with independent motions. We first show that often objects moving with different 3D motions will be captured as a single object using previously suggested approaches.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

On Single-Sequence and Multi-Sequence Factorizations

Zelnik-Manor

Irani

2006

Int J Comput Vision

View full text Add to dashboard Cite

show abstract

Section: Figure 1 Degeneracies In Multi-body (Single Sequence) Factomentioning

confidence: 99%

Section: Figure 1 Degeneracies In Multi-body (Single Sequence) Factomentioning

confidence: 99%

See 1 more Smart Citation

On Single-Sequence and Multi-Sequence Factorizations

Zelnik-Manor

Irani

2006

Int J Comput Vision

View full text Add to dashboard Cite

show abstract

“…The second stage often is performed through factorization methods [2], [5], [15], [26], although some simpler clustering strategy can be used [27], [28]. In factorization methods, motion and shape information are treated separately by applying constraints to the scene projection on the image formation plane, as well as on the object shape and motion.…”

mentioning

confidence: 99%

Video Segmentation Based on Motion Coherence of Particles in a Video Sequence

Silva

Scharcanski

2010

IEEE Trans. on Image Process.

View full text Add to dashboard Cite

Abstract-This work describes an approach for object-oriented video segmentation based on motion coherence. Using a tracking process based on adaptively sampled points (namely, particles), 2-D motion patterns are identified with an ensemble clustering approach. Particles are clustered to obtain a pixel-wise segmentation in space and time domains. The segmentation result is mapped to an image spatio-temporal feature space. Thus, the different constituent parts of the scene that move coherently along the video sequence are mapped to volumes in this spatio-temporal space. These volumes make the redundancy in the temporal sense more explicit, leading to potential gains in video coding applications. The proposed solution is robust and more generic than similar approaches for 2-D video segmentation found in the literature. In order to illustrate the potential advantages of using the proposed motion segmentation approach in video coding applications, the PSNR of the temporal predictions and the entropies of prediction errors obtained in our experiments are presented, and compared with other methods. Our experiments with real and synthetic sequences suggest that our method also could be used in other image processing and computer vision tasks, besides video coding, such as video information retrieval and video understanding.

show abstract

“…The body of work on multi-body factorization (e.g., [3,4,5,1,7]), suggest employing multi-frame linear subspace constraints to separate between objects moving with independent motions. We first show that often objects moving with different 3D motions will be captured as a single object using previously suggested approaches.…”

Section: Introductionmentioning

confidence: 99%