Multi-Task Recurrent Neural Network for Immediacy Prediction
Abstract
In this paper, we propose to predict immediacy for interacting persons from still images. A complete immediacy set includes interactions, relative distance, body leaning direction and standing orientation. These measures are found to be related to the attitude, social relationship, social interaction, action, nationality, and religion of the communicators. A large-scale dataset with 10,000 images is constructed, in which all the immediacy measures and the human poses are annotated. We propose a rich set of immediacy representations that help to predict immediacy from imperfect 1-person and 2-person pose estimation results. A multi-task deep recurrent neural network is constructed to take the proposed rich immediacy representation as input and learn the complex relationship among immediacy predictions multiple steps of refinement. The effectiveness of the proposed approach is proved through extensive experiments on the large scale dataset.
Cite
Text
Chu et al. "Multi-Task Recurrent Neural Network for Immediacy Prediction." International Conference on Computer Vision, 2015. doi:10.1109/ICCV.2015.383Markdown
[Chu et al. "Multi-Task Recurrent Neural Network for Immediacy Prediction." International Conference on Computer Vision, 2015.](https://mlanthology.org/iccv/2015/chu2015iccv-multitask/) doi:10.1109/ICCV.2015.383BibTeX
@inproceedings{chu2015iccv-multitask,
title = {{Multi-Task Recurrent Neural Network for Immediacy Prediction}},
author = {Chu, Xiao and Ouyang, Wanli and Yang, Wei and Wang, Xiaogang},
booktitle = {International Conference on Computer Vision},
year = {2015},
doi = {10.1109/ICCV.2015.383},
url = {https://mlanthology.org/iccv/2015/chu2015iccv-multitask/}
}