Places205-VGGNet Models for Scene Recognition
Published in
arXiv.org(2015)External Links:
TL
TL;DR
This report describes the implementation of training the VGGNets on the large-scale Places205 dataset by using a Multi-GPU extension of Caffe toolbox with high computational efficiency and achieves the state-of-the-art performance of trained models on three datasets.
Abstract
VGGNets have turned out to be effective for object recognition in still images. However, it is unable to yield good performance by directly adapting the VGGNet models trained on the ImageNet dataset for scene recognition. This report describes our implementation of training the VGGNets on the large-scale Places205 dataset. Specifically, we train three VGGNet models, namely VGGNet-11, VGGNet-13, and VGGNet-16, by using a Multi-GPU extension of Caffe toolbox with high computational efficiency. We verify the performance of trained Places205-VGGNet models on three datasets: MIT67, SUN397, and Places205. Our trained models achieve the state-of-the-art performance o these datasets and are made public available 1 .
References15 items
1
Towards Good Practices for Very Deep Two-Stream ConvNets
2
Training Deeper Convolutional Networks with Deep Supervision
3
Learning Deep Features for Scene Recognition using Places Database
4
Going deeper with convolutions
5