A contour detection pipeline using H-ELM and DenseCRF is reported to outperform SVM, ELM, and a simple CNN on a 200-image Cassini dataset, with about 9 seconds of training time.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
astro-ph.IM 1years
2019 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Contour Detection in Cassini ISS images based on Hierarchical Extreme Learning Machine and Dense Conditional Random Field
A contour detection pipeline using H-ELM and DenseCRF is reported to outperform SVM, ELM, and a simple CNN on a 200-image Cassini dataset, with about 9 seconds of training time.