Segmentation > Superpixels
Groups pixels into regions with similar values. This can be a helpful processing step prior to other segmentation or feature-finding steps. Uses the simple linear iterative clustering (SLIC) algorithm [1,2].
1. Type
Statistic to use to fill each cluster
- Mean: Fills each superpixel with mean value of underlying pixels
- Median: Fills each superpixel with median value of underlying pixels
- Mode: Fills each superpixel with mode value of underlying pixels
- StdDev: Fills each superpixel with standard deviation of underlying pixels
- Min: Fills each superpixel with minimum value of underlying pixels
- Max: Fills each superpixel with maximum value of underlying pixels
2. Number
The target number of superpixels to create. Actual number may be different.
3. Shape
Factor which controls the shape of the superpixels. Lower numbers produce more realistic shapes that follow boundaries. (Recommended: 10)
References
[1] Radhakrishna Achanta, Appu Shaji, Kevin Smith, Aurelien Lucchi, Pascal Fua, and Sabine Susstrunk, SLIC Superpixels Compared to State-of-the-art Superpixel Methods. IEEE Transactions on Pattern Analysis and Machine Intelligence, Volume 34, Issue 11, pp. 2274-2282, May 2012
[2] Radhakrishna Achanta, Appu Shaji, Kevin Smith, Aurelien Lucchi, Pascal Fua, and Sabine Susstrunk, SLIC Superpixels. EPFL Technical Report, no. 149300, June 2010.
Need more help with this?
Chat with an expert now ››