| Grant number: | 15/01186-6 |
| Support Opportunities: | Research Grants - Visiting Researcher Grant - International |
| Start date: | June 16, 2015 |
| End date: | July 31, 2015 |
| Field of knowledge: | Physical Sciences and Mathematics - Computer Science - Computing Methodologies and Techniques |
| Principal Investigator: | Alexandre Xavier Falcão |
| Grantee: | Alexandre Xavier Falcão |
| Visiting researcher: | Ananda Shankar Chowdhury |
| Visiting researcher institution: | Jadavpur University, India |
| Host Institution: | Instituto de Computação (IC). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil |
Abstract
Superpixel based research has gained much importance in the field of computer vision over the last decade. Results from algorithms based on superpixels show promise in typical computer vision problems like segmentation, tracking, and, depth estimation. However, generation of superpixels still remains a challenging problem. Many graph-based as well as gradient ascent-based methods are applied in the past for segmentation of superpixels. Lately, SLIC approach for superpixel generation has yielded better performance than the other existing techniques. The main objective of this proposal is to improve the SLIC based segmentation of superpixels by replacing the k-means clustering algorithm with the Optimum Path Forest (OPF) method. Some theoretical improvements of the OPF algorithm will also be investigated, especially for large datasets, to further enhance the quality of the superpixels. Finally, we will show improved segmentation performances on medical images as well as on videos with the newly generated improved superpixels. (AU)
| Articles published in Agência FAPESP Newsletter about the research grant: |
| More itemsLess items |
| TITULO |
| Articles published in other media outlets ( ): |
| More itemsLess items |
| VEICULO: TITULO (DATA) |
| VEICULO: TITULO (DATA) |