Electronic Thesis/Dissertation
Differential GradCAM for CNN-based Fine-Grained Visual Categorization Model
Open AccessFine-grained classification requires a visual interpretation method that can capture subtle differences between similar classes, which classical approaches may not be able to achieve. In this paper, we introduce a new approach called Differential Grad-CAM, which can effectively localize differences between pairs of similar classes. This method helps us better understand the CNN-based models used in fine-grained classification and identify new regions for visual explanation. Differential Grad-CAM provides a new perspective for interpreting CNN-based models in fine-grained classification.
- All rights reserved
Notice to Authors
If you are the author of this work and you have any questions about the information on this page, please use the Contact form to get in touch with us.