Stating the Obvious: Extracting Visual Common Sense Knowledge

@inproceedings{Yatskar2016StatingTO,
  title={Stating the Obvious: Extracting Visual Common Sense Knowledge},
  author={Mark Yatskar and Vicente Ordonez and Ali Farhadi},
  booktitle={HLT-NAACL},
  year={2016}
}
Obtaining common sense knowledge using current information extraction techniques is extremely challenging. In this work, we instead propose to derive simple common sense statements from fully annotated object detection corpora such as the Microsoft Common Objects in Context dataset. We show that many thousands of common sense facts can be extracted from such corpora at high quality. Furthermore, using WordNet and a novel submodular k-coverage formulation, we are able to generalize our initial… CONTINUE READING

Citations

Publications citing this paper.
SHOWING 1-10 OF 22 CITATIONS

Unambiguous Scene Text Segmentation With Referring Expression Comprehension

VIEW 1 EXCERPT
CITES BACKGROUND

References

Publications referenced by this paper.
SHOWING 1-10 OF 24 REFERENCES

WordNet: a lexical database for English

VIEW 7 EXCERPTS
HIGHLY INFLUENTIAL

Predicting Entry-Level Categories

VIEW 2 EXCERPTS

VisKE: Visual knowledge extraction and question answering by visual verification of relation phrases

VIEW 2 EXCERPTS

Learning Everything about Anything: Webly-Supervised Visual Concept Learning

VIEW 2 EXCERPTS