Morphological shape context : semi-locality and robust matching in shape recognition
Resumen:
We present a novel shape recognition method based on an al-gorithm to detect contrasted level lines for extraction, on Shape Context for encoding and on an a contrario approach for matching. The contri-butions naturally lead to a semi-local Shape Context. Results show that this method is able to work in contexts where Shape Context cannot, such as content-based video retrieval.
| 2009 | |
|
Level line Canny edge detector Shape context Maximal stable extremal region Jitter noise |
|
| Inglés | |
| Universidad de la República | |
| COLIBRI | |
| https://hdl.handle.net/20.500.12008/38685 | |
| Acceso abierto | |
| Licencia Creative Commons Atribución - No Comercial - Sin Derivadas (CC - By-NC-ND 4.0) |
| Sumario: | We present a novel shape recognition method based on an al-gorithm to detect contrasted level lines for extraction, on Shape Context for encoding and on an a contrario approach for matching. The contri-butions naturally lead to a semi-local Shape Context. Results show that this method is able to work in contexts where Shape Context cannot, such as content-based video retrieval. |
|---|
