Open Access
Table 2
Evaluation Metrics for Different Methods. Lower values for Wasserstein Distance and Jensen–Shannon Divergence indicate better performance, while a higher value for Divergence Amongst Generated Images signifies greater diversity.
| Method | Wasserstein Distance | Jensen–Shannon Divergence | Image Diversity |
|---|---|---|---|
| Proposed Method | 0.27 | 0.35 | 0.47 |
| Proposed Method + ADA | 0.41 | 0.54 | 0.84 |
| Affine Transformation | 0.21 | 0.56 | 0.58 |
| Erasing Transformation | 1.21 | 0.61 | 0.73 |
| Pixel level Transformation | 0.11 | 0.49 | 0.69 |
The bold values indicate that the best results were achieved under those specific conditions.
Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.
Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.
Initial download of the metrics may take a while.