Conference paper Open Access

Comparison of Fine-tuning and Extension Strategies for Deep Convolutional Neural Networks

Pittaras, Nikiforos; Markatopoulou, Foteini; Mezaris, Vasileios; Patras, Ioannis

Citation Style Language JSON Export

{
"DOI": "10.1007/978-3-319-51811-4_9",
"author": [
{
"family": "Pittaras, Nikiforos"
},
{
"family": "Markatopoulou, Foteini"
},
{
"family": "Mezaris, Vasileios"
},
{
"family": "Patras, Ioannis"
}
],
"issued": {
"date-parts": [
[
2016,
12,
31
]
]
},
"abstract": "<p>In this study we compare three different fine-tuning strategies in order to investigate the best way to transfer the parameters of popular deep convolutional neural networks that were trained for a visual annotation task on one dataset, to a new, considerably different dataset. We focus on the concept-based image/video annotation problem and use ImageNet as the source dataset, while the TRECVID SIN 2013 and PASCAL VOC-2012 classification datasets are used as the target datasets. A large set of experiments examines the effectiveness of three fine-tuning strategies on each of three different pre-trained DCNNs and each target dataset. The reported results give rise to guidelines for effectively fine-tuning a DCNN for concept-based visual annotation.</p>",
"title": "Comparison of Fine-tuning and Extension Strategies for Deep Convolutional Neural Networks",
"type": "paper-conference",
"id": "240853"
}
132
103
views
downloads
 Views 132 Downloads 103 Data volume 26.3 MB Unique views 127 Unique downloads 98