Corpus ID: 31728242

A Cost-Sensitive Visual Question-Answer Framework for Mining a Deep And-OR Object Semantics from Web Images

@article{Zhang2017ACV,
  title={A Cost-Sensitive Visual Question-Answer Framework for Mining a Deep And-OR Object Semantics from Web Images},
  author={Q. Zhang and Y. Wu and S. Zhu},
  journal={ArXiv},
  year={2017},
  volume={abs/1708.03911}
}
  • Q. Zhang, Y. Wu, S. Zhu
  • Published 2017
  • Computer Science
  • ArXiv
  • This paper presents a cost-sensitive Question-Answering (QA) framework for learning a nine-layer And-Or graph (AoG) from web images, which explicitly represents object categories, poses, parts, and detailed structures within the parts in a compositional hierarchy. The QA framework is designed to minimize an overall risk, which trades off the loss and query costs. The loss is defined for nodes in all layers of the AoG, including the generative loss (measuring the likelihood for the images) and… CONTINUE READING

    Figures, Tables, and Topics from this paper.

    Explanatory Graphs for CNNs
    Interpreting CNN knowledge via an Explanatory Graph
    81
    Knowledge Via An Explanatory Graph
    1
    Development and Challenges of Phenotypic Characterization in Modal Animals

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 49 REFERENCES
    VQA: Visual Question Answering
    1753
    Strong supervision from weak annotation: Interactive training of deformable part models
    105
    Joint Video and Text Parsing for Understanding Events and Answering Queries
    95
    Best of both worlds: Human-machine collaboration for object annotation
    131
    Enriching Visual Knowledge Bases via Object Discovery and Segmentation
    110
    What Are You Talking About? Text-to-Image Coreference
    146