File Download
There are no files associated with this item.
Supplementary
-
Citations:
- Appears in Collections:
Conference Paper: Enhanced Hierarchical Conditional Random Field Model for Semantic Image Segmentation
Title | Enhanced Hierarchical Conditional Random Field Model for Semantic Image Segmentation |
---|---|
Authors | |
Issue Date | 2014 |
Citation | The 9th International Conference on Computer Vision Theory and Applications (VISAPP), Lisbon, Portugal, 5-8 January 2014, p. 215-222 How to Cite? |
Abstract | Pairwise and higher order potentials in the Hierarchical Conditional Random Field (HCRF) model play a vital role in smoothing region boundary and extracting actual object contour in the labeling space. However, pairwise potential evaluated by color information has the tendency to over-smooth small regions which are similar to their neighbors in the color space; and the higher order potential associated with multiple segments is prone to produce incorrect guidance to inference, especially for objects having similar features to the background. To overcome these problems, this paper proposes two enhanced potentials in the HCRF model that is capable to abate the over smoothness by propagating the believed labeling from the unary potential and to perform coherent inference by ensuring reliable segment consistency. Experimental results on the MSRC-21 data set demonstrate that the enhanced HCRF model achieves pleasant visual results, as well as significant improvement in terms of both global accuracy of 87.52% and average accuracy of 80.18%, which outperforms other algorithms reported in the literature so far. |
Description | Area 3 - Image and Video Understanding Short paper: paper no. 16 |
Persistent Identifier | http://hdl.handle.net/10722/204075 |
DC Field | Value | Language |
---|---|---|
dc.contributor.author | Wang, L | en_US |
dc.contributor.author | Zhu, S | en_US |
dc.contributor.author | Yung, NHC | en_US |
dc.date.accessioned | 2014-09-19T20:04:22Z | - |
dc.date.available | 2014-09-19T20:04:22Z | - |
dc.date.issued | 2014 | en_US |
dc.identifier.citation | The 9th International Conference on Computer Vision Theory and Applications (VISAPP), Lisbon, Portugal, 5-8 January 2014, p. 215-222 | en_US |
dc.identifier.uri | http://hdl.handle.net/10722/204075 | - |
dc.description | Area 3 - Image and Video Understanding | - |
dc.description | Short paper: paper no. 16 | - |
dc.description.abstract | Pairwise and higher order potentials in the Hierarchical Conditional Random Field (HCRF) model play a vital role in smoothing region boundary and extracting actual object contour in the labeling space. However, pairwise potential evaluated by color information has the tendency to over-smooth small regions which are similar to their neighbors in the color space; and the higher order potential associated with multiple segments is prone to produce incorrect guidance to inference, especially for objects having similar features to the background. To overcome these problems, this paper proposes two enhanced potentials in the HCRF model that is capable to abate the over smoothness by propagating the believed labeling from the unary potential and to perform coherent inference by ensuring reliable segment consistency. Experimental results on the MSRC-21 data set demonstrate that the enhanced HCRF model achieves pleasant visual results, as well as significant improvement in terms of both global accuracy of 87.52% and average accuracy of 80.18%, which outperforms other algorithms reported in the literature so far. | - |
dc.language | eng | en_US |
dc.relation.ispartof | International Conference on Computer Vision Theory and Applications (VISAPP) | en_US |
dc.title | Enhanced Hierarchical Conditional Random Field Model for Semantic Image Segmentation | en_US |
dc.type | Conference_Paper | en_US |
dc.identifier.email | Wang, L: llwang@hku.hk | en_US |
dc.identifier.email | Yung, NHC: nyung@eee.hku.hk | en_US |
dc.identifier.authority | Yung, NHC=rp00226 | en_US |
dc.identifier.hkuros | 238537 | en_US |
dc.identifier.spage | 215 | en_US |
dc.identifier.epage | 222 | en_US |