Abstract
Recognizing objects in natural images is an intricate problem involving multiple conflicting objectives. Deep convolutional neural networks, trained on large datasets, achieve convincing results and are currently the state-of-the-art approach for this task. However, the long time needed to train such deep networks is a major drawback. We tackled this problem by reusing a previously trained network. For this purpose, we first trained a deep convolutional network on the ILSVRC-12 dataset. We then maintained the learned convolution kernels and only retrained the classification part on different datasets. Using this approach, we achieved an accuracy of 67.68% on CIFAR-100, compared to the previous state-of-the-art result of 65.43%. Furthermore, our findings indicate that convolutional networks are able to learn generic feature extractors that can be used for different tasks.
Original language | English |
---|---|
Title of host publication | 2015 International Joint Conference on Neural Networks (IJCNN) |
Number of pages | 4 |
Publisher | IEEE |
Publication date | 28.09.2015 |
ISBN (Print) | 978-1-4799-1961-1 |
ISBN (Electronic) | 978-1-4799-1960-4 |
DOIs | |
Publication status | Published - 28.09.2015 |
Event | International Joint Conference on Neural Networks 2015 - Killarney, Ireland Duration: 12.07.2015 → 17.07.2015 https://ieeexplore.ieee.org/document/7280683/ |