Motion estimation with detection of occlusion areas

Depommier, R.; Dubois, Éric

doi:10.1109/icassp.1992.226199

Cited by 25 publications

(16 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To penalize the introduction of a label, in addition to the horizontal and vertical cliques from Figure 3.3, single-element cliques are used as well. For the 3-state occlusion field presented in Table 3.1(a), the choice of potential values is similar to the one used in [13] and is shown in Figure 3.4. To ensure that occlusion states get clustered, V o = 0 (high probability) for adjacent identical labels, and a high value of V o (low probability) for different labels should be specified ( Figure 3.4(b)).…”

Section: Occlusion Modelmentioning

confidence: 99%

“…It should be noted that errors in motion estimates can lead to erroneously interpolated values that can seriously impair the picture, thus making highly reliable motion estimates a necessity for motion-compensated interpolation. If the field to be interpolated is available at the motion estimation phase (as in some coding systems where the motion field is transmitted) much better interpolation results can often be obtained [13].…”

Section: Motion-compensated Interpolationmentioning

confidence: 99%

“…The overall energy function can be a linear combination of (3.23) and (3.24): 25) where γ must be determined according to the structural model. Previous work has used γ = 1 and α i = 0, i = 1, ..., K [13].…”

Section: Structural Model and Matching Errormentioning

confidence: 99%

“…Similarly, if I x t = I t , the potential V (3) g can be set to zero, since the two-element clique remaining is sufficient. Thus, a general expression for the energy function is [13] …”

Section: Structural Model and Matching Errormentioning

confidence: 99%

“…In typical images, motion fields are usually smooth functions of spatial position x, except at motion boundaries. Thus, we model trajectories c p t by continuous-valued vector MRFs C p t [46], [35], and also we model occlusions o t and motion discontinuities l t by multi-level and binary MRFs O t [13] and L t [28], [37], respectively. We ensure the smoothness of c p t and o t , and continuity of l t by appropriate choice of Gibbs distribution parameters.…”

Section: Motion Modelsmentioning

confidence: 99%

See 4 more Smart Citations

Estimation of 2-D Motion Fields from Image Sequences with Application to Motion-Compensated Processing

Dubois¹,

Konrad²

1993

Motion Analysis and Image Sequence Processing

Self Cite

View full text Add to dashboard Cite

In this chapter we are concerned with the estimation of 2-D motion from time-varying images and with the application of the computed motion to image sequence processing. Our goal for motion estimation is to propose a general formulation that incorporates object acceleration, nonlinear motion trajectories, occlusion effects and multichannel (vector) observations. To achieve this objective we use Gibbs-Markov models linked together by the Maximum A Posteriori Probability criterion which results in minimization of a multiple-term cost function. The specific applications of motion-compensated processing of image sequences are prediction, noise reduction and spatiotemporal interpolation.Estimation of motion from dynamic images is a very difficult task due to its ill-posedness [4]. Despite this difficulty, however, many approaches to the problem have been proposed in the last dozen years [27], [24], [40]. This activity can certainly be attributed in large measure to the importance of motion in the processing and coding of image sequences. Below we explain why motion is important in these tasks.Efficient encoding of time-varying images is essential to provide economical use of network or storage facilities in the provision of video services. Image sequences can be compressed by independent coding of each frame (intraframe coding) or by straightforward extension of spatial coding techniques to three dimensions (e.g., 3-D transform coding). However, such approaches ignore the fact that the majority of new information (innovations) in a time-varying image is carried by the motion. The correlation of image intensity or color is very high along the direction of motion. Thus, the knowledge of motion helps in removing significant interimage redundancy, as is the case in predictive or hybrid (predictive/transform or predictive/subband) coding compensated for motion. In fact, these techniques include most algorithms currently used in videoconferencing [41], [42] and proposed for High Definition Television (HDTV) [21], [60].Sampling structure conversion and noise reduction are two examples of image sequence processing that can benefit greatly from the knowledge of motion. In the case of conversion, two scenarios are possible. In one situation a missing image must be recovered. Again, due to the high correlation along motion trajectories, motion-compensated interpolation is the most effective tool [53], [15]. In the other

show abstract

Section: Occlusion Modelmentioning

confidence: 99%

Section: Motion-compensated Interpolationmentioning

confidence: 99%

Section: Structural Model and Matching Errormentioning

confidence: 99%

“…Similarly, if I x t = I t , the potential V (3) g can be set to zero, since the two-element clique remaining is sufficient. Thus, a general expression for the energy function is [13] …”

Section: Structural Model and Matching Errormentioning

confidence: 99%

Section: Motion Modelsmentioning

confidence: 99%

See 3 more Smart Citations

Estimation of 2-D Motion Fields from Image Sequences with Application to Motion-Compensated Processing

Dubois¹,

Konrad²

1993

Motion Analysis and Image Sequence Processing

Self Cite

View full text Add to dashboard Cite

show abstract

Gibbs Model Based 3D Motion and Structure Estimation for Object-Based Video Coding Applications

Alatan

Onural

1997

Video Data Compression for Multimedia Computing

View full text Add to dashboard Cite

Motion analysis is essential for any video coding scheme. A moving object in a 3D environment can be analyzed better by a 3D motion model instead of 2D models, and better modeling might lead to improved coding efficiency. Gibbs formulated joint segmentation and estimation of 2D motion not only improves the performance of each stage, but also generates robust point correspondences which are necessary for rigid 3D motion estimation algorithms. Estimated rigid 3D motion parameters of a segmented object are used to find the 3D structure of those objects by minimizing another Gibbs energy. Such an approach achieves error immunity compared to linear algorithms. A more general (non-rigid) motion model can also be proposed using Gibbs formulation which permits local elastic interactions in contrast to ultimately tight rigidity between object points. Experimental results are promising for both rigid and non-rigid 3D motion models and put these models forward as strong candidates to be used in object-based coding algorithms.

show abstract

Estimation and compensation of accelerated motion for temporal sequence interpolation

Chahine

Konrad

1995

Signal Processing: Image Communication

View full text Add to dashboard Cite

Motion estimation with detection of occlusion areas

Cited by 25 publications

References 10 publications

Estimation of 2-D Motion Fields from Image Sequences with Application to Motion-Compensated Processing

Estimation of 2-D Motion Fields from Image Sequences with Application to Motion-Compensated Processing

Gibbs Model Based 3D Motion and Structure Estimation for Object-Based Video Coding Applications

Estimation and compensation of accelerated motion for temporal sequence interpolation

Contact Info

Product

Resources

About