| Full text | |
| Author(s): |
Vieira, Miguel G.
;
Moreira, Jander
;
GarciaGoncalves, LM
;
BeserraGomes, R
Total Authors: 4
|
| Document type: | Journal article |
| Source: | 2017 WORKSHOP OF COMPUTER VISION (WVC); v. N/A, p. 6-pg., 2017-01-01. |
| Abstract | |
The amount of images we find online nowadays is huge. Therefore the still standing interest in the research in Image Classification, seeking ways to better classify these images. One of those kinds of researches is using Neural Networks to better classify those images. Another line of research is using Hierarchical Classifiers to differentiate the content of the images. In this paper, we unite both types of classification, using Hierarchical Classifier trained using Neural Networks, to after, compare the results to both of classifications alone. The results we obtained showed a significant improvement in comparison to traditional methods. The research was focused on e-commerce, since all the images used to train the classifier were obtained from products easily found in e-stores. (AU) | |
| FAPESP's process: | 16/13002-0 - MMeaning - multimodal distributional semantic models |
| Grantee: | Helena de Medeiros Caseli |
| Support Opportunities: | Regular Research Grants |