Combining Multi-level Contexts of Superpixel using Convolutional Neural Networks to perform Natural Scene Labeling
This work addresses image segmentation for natural scenes by leveraging superpixels to reduce computation and incorporate context, though it appears incremental as it builds on existing deep learning and superpixel methods.
The paper tackles natural scene labeling by performing superpixel-level semantic segmentation using multi-level neighbor contexts and ensemble methods, achieving superior performance compared to modern approaches on the same dataset.
Modern deep learning algorithms have triggered various image segmentation approaches. However most of them deal with pixel based segmentation. However, superpixels provide a certain degree of contextual information while reducing computation cost. In our approach, we have performed superpixel level semantic segmentation considering 3 various levels as neighbours for semantic contexts. Furthermore, we have enlisted a number of ensemble approaches like max-voting and weighted-average. We have also used the Dempster-Shafer theory of uncertainty to analyze confusion among various classes. Our method has proved to be superior to a number of different modern approaches on the same dataset.