An Empirical Evaluation of Visual Question Answering for Novel Objects

@article{Ramakrishnan2017AnEE,
  title={An Empirical Evaluation of Visual Question Answering for Novel Objects},
  author={Santhosh K. Ramakrishnan and Ambar Pal and Gaurav Sharma and Anurag Mittal},
  journal={2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
  year={2017},
  pages={7312-7321}
}
We study the problem of answering questions about images in the harder setting, where the test questions and corresponding images contain novel objects, which were not queried about in the training data. Such setting is inevitable in real world–owing to the heavy tailed distribution of the visual categories, there would be some objects which would not be annotated in the train set. We show that the performance of two popular existing methods drop significantly (21–28%) when… CONTINUE READING
Recent Discussions
This paper has been referenced on Twitter 10 times over the past 90 days. VIEW TWEETS

References

Publications referenced by this paper.
Showing 1-10 of 42 references

Similar Papers

Loading similar papers…