in illumination or occlusion (and by other moving objects). In fact, there have been a number of
studies on performance, e.g. of affine flow in Grossmann and Santos-Victor (1997). A thorough
analysis of correlation techniques has been developed (Giachetti, 2000) with new algorithms for
sub-pixel estimation. One study (Liu et al., 1998) notes how developments have been made for
fast or for accurate techniques, without consideration of the trade-off between these two factors.
The study compared the techniques mentioned previously with two newer approaches (one fast
and one accurate), and also surveys real-time implementations that include implementation via
parallel computers and special purpose VLSI chips.
4.10 Conclusions
This chapter has covered the main ways to extract low-level feature information. In some cases
this can prove sufficient for understanding the image. Often, however, the function of low-level
feature extraction is to provide information for later higher level analysis. This can be achieved
in a variety of ways, with advantages and disadvantages and quickly or at a lower speed (or
requiring a faster processor/more memory!). The