Abstract
The object masks obtained from the spatial-temporal segmentation procedure described in Section III sometimes have irregularity in extracted boundaries such as small gulfs and isthmi due to temporal and spatial signal fluctuations. These irregular features lead to a visually annoying appearance and also increase the cost of shape coding. We adopt two boundary postprocessing techniques to handle this problem. First, we use the morphological open and close operators to remove unwanted gulfs and isthmi for object boundary smoothing, where a circular structuring element with a radius of 2 pixels is used in these morphological filters. Second, we consider the confidence level of extracted object boundaries by using boundaries of variable width, where a higher (or lower) confidence level leads to a narrower (or wider) boundary width. Consequently, we are able to adjust the shape of the segmentation mask in low confidence regions more freely to increase the shape coding efficiency.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2002 Springer Science+Business Media New York
About this chapter
Cite this chapter
Guo, J., Kuo, CC.J. (2002). Object Shape Postprocessing. In: Semantic Video Object Segmentation for Content-Based Multimedia Applications. The Springer International Series in Engineering and Computer Science, vol 639. Springer, Boston, MA. https://doi.org/10.1007/978-1-4615-1503-6_4
Download citation
DOI: https://doi.org/10.1007/978-1-4615-1503-6_4
Publisher Name: Springer, Boston, MA
Print ISBN: 978-1-4613-5586-1
Online ISBN: 978-1-4615-1503-6
eBook Packages: Springer Book Archive