Estimation of occlusion and dense motion fields in a bidirectional Bayesian framework

IEEE Transactions on Pattern Analysis and Machine Intelligence

Volumn 24, Issue 5, 2002, Pages 712-718

  Lim, Keng Pang ^a,b   Das, Amitabha ^a,c   Chong, Man Nang ^a,d  

^a IEEE   (Singapore)

^b NANYANG TECHNOLOGICAL UNIVERSITY   (Singapore)

^c NANYANG TECHNOLOGICAL UNIVERSITY   (Singapore)

^d GDC Technology Limited   (Hong Kong)

Author keywords

[No Author keywords available]

Indexed keywords

BIDIRECTIONAL BAYESIAN FRAMEWORK; BIDIRECTIONAL RELAXATION; DENSE MOTION FIELD ESTIMATION; DUALITY PRINCIPLES; ITERATED CONDITIONAL MODE RELAXATION; MARKOV RANDOM FIELDS; MAXIMUM A POSTERIORI PROBABILITY ESTIMATION; MOTION SEQUENCE; MOTION VECTOR; OCCLUSION DETECTION;

ALGORITHMS; CALCULATIONS; IMAGE ANALYSIS; IMAGING TECHNIQUES; MARKOV PROCESSES; MATHEMATICAL MODELS; MOTION COMPENSATION; OPTIMIZATION; VECTORS;

MOTION ESTIMATION;

EID: 0036565804     PISSN: 01628828     EISSN: None     Source Type: Journal    
DOI: 10.1109/34.1000246     Document Type: Article

References (14)

1. Estimation of 2-D fields from image sequences with application to motion-compensated processing
2. The application of mean field theory to image motion estimation
3. Spatial interaction and the statistical analysis of lattice systems
4. Image segmentation based on object oriented mapping parameter estimation
5. Motion estimation with detection of occlusion areas
6. Multimodal estimation of discontinous optical flow using markov random fields
7. Estimation of accelerated motion and occlusions from time-varying images
8. Bayesian estimation of motion vector fields
9. Correspondence estimation in image pairs
10. The robust estimation of multiple motions parametric and piecewise-smooth flow fields
11. Motion compensating interpolation considering covered and uncovered background
12. A bayes decision test for detecting uncovered background and moving pixels in image sequences