As mentioned earlier, the labels in the segmentation problem are one-dimensional images with a value at each pixel that matches the index of the output volume depth:
Labels
by Leonardo Araujo, Nimesh Patel, Richard Burton, Giounona Tzanidou, Iffat Zafar
Hands-On Convolutional Neural Networks with TensorFlow
- Title Page
- Copyright and Credits
- Packt Upsell
- Contributors
- Preface
- Setup and Introduction to TensorFlow
- Deep Learning and Convolutional Neural Networks
- Image Classification in TensorFlow
- Object Detection and Segmentation
- VGG, Inception Modules, Residuals, and MobileNets
- Autoencoders, Variational Autoencoders, and Generative Adversarial Networks
- Transfer Learning
- Machine Learning Best Practices and Troubleshooting
- Training at Scale
- References
- Other Books You May Enjoy
Labels
-
No Comment
..................Content has been hidden....................
You can't read the all page of ebook, please click here login for view all page.