Ideal Observers for Detecting Motion: Correspondence Noise

@inproceedings{Lu2005IdealOF,
  title={Ideal Observers for Detecting Motion: Correspondence Noise},
  author={Hongjing Lu and Alan L. Yuille},
  booktitle={NIPS},
  year={2005}
}
We derive a Bayesian Ideal Observer (BIO) for detecting moti on and solving the correspondence problem. We obtain Barlow and Tr ipathy’s classic model as an approximation. Our psychophysical expe riments show that the trends of human performance are similar to the B ayesian Ideal, but overall human performance is far worse. We invest igate ways to degrade the Bayesian Ideal but show that even extreme degr adations do not approach human performance. Instead we propose that h umans perform motion… CONTINUE READING
Highly Cited
This paper has 33 citations. REVIEW CITATIONS

Citations

Publications citing this paper.
Showing 1-10 of 19 extracted citations

References

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

The efficiency of depth dis crimination for non-transparent and transparent stereoscopic surfaces

  • Wallace, J.M, P. Mamassian
  • Vision Research,
  • 2004
1 Excerpt

The interpretation of Visual Motion

  • S. Ullman
  • MIT Press, C ambridge,
  • 1979
1 Excerpt

Similar Papers

Loading similar papers…