Dynamically Identifying Deep Multimodal Features for Image Privacy Prediction

AAAI 2019 pp. 10057-10058

doi:10.1609/AAAI.V33I01.330110057 /aaai/2019/tonge2019aaai-dynamically/

Abstract

With millions of images shared online, privacy concerns are on the rise. In this paper, we propose an approach to image privacy prediction by dynamically identifying powerful features corresponding to objects, scene context, and image tags derived from Convolutional Neural Networks for each test image. Specifically, our approach identifies the set of most “competent” features on the fly, according to each test image whose privacy has to be predicted. Experimental results on thousands of Flickr images show that our approach predicts the sensitive (or private) content more accurately than the models trained on each individual feature set (object, scene, and tags alone) or their combination.

PDF AAAI Semantic Scholar

Cite

Text

Tonge and Caragea. "Dynamically Identifying Deep Multimodal Features for Image Privacy Prediction." AAAI Conference on Artificial Intelligence, 2019. doi:10.1609/AAAI.V33I01.330110057

Markdown

[Tonge and Caragea. "Dynamically Identifying Deep Multimodal Features for Image Privacy Prediction." AAAI Conference on Artificial Intelligence, 2019.](https://mlanthology.org/aaai/2019/tonge2019aaai-dynamically/) doi:10.1609/AAAI.V33I01.330110057

BibTeX

@inproceedings{tonge2019aaai-dynamically,
  title     = {{Dynamically Identifying Deep Multimodal Features for Image Privacy Prediction}},
  author    = {Tonge, Ashwini and Caragea, Cornelia},
  booktitle = {AAAI Conference on Artificial Intelligence},
  year      = {2019},
  pages     = {10057-10058},
  doi       = {10.1609/AAAI.V33I01.330110057},
  url       = {https://mlanthology.org/aaai/2019/tonge2019aaai-dynamically/}
}