Where Should Saliency Models Look Next?

  title={Where Should Saliency Models Look Next?},
  author={Zoya Bylinskii and Adri{\`a} Recasens and Ali Borji and Aude Oliva and Antonio Torralba and Fr{\'e}do Durand},
Pr ed ic tio ns Have saliency models begun to converge on human performance? We re-examine the current state-ofthe-art using a fine-grained analysis on image types, individual images, and image regions. We quantify up to 60% of remaining errors of saliency models. To continue to approach human-level performance, saliency models will need to discover higher-level concepts in images and reason about the relative importance of image regions. 
Highly Cited
This paper has 77 citations. REVIEW CITATIONS

From This Paper

Topics from this paper.


Publications citing this paper.
Showing 1-10 of 46 extracted citations

78 Citations

Citations per Year
Semantic Scholar estimates that this publication has 78 citations based on the available data.

See our FAQ for additional information.


Publications referenced by this paper.
Showing 1-10 of 34 references

Similar Papers

Loading similar papers…