Optical flow

Consider a five-frame clip of a ball moving from the bottom left of a field of vision, to the top right. Motion estimation techniques can determine that on a two dimensional plane the ball is moving up and to the right and vectors describing this motion can be extracted from the sequence of frames. For the purposes of video compression (e.g., [[MPEG]]), the sequence is now described as well as it needs to be. However, in the field of machine vision, the question of whether the ball is moving to the right or if the observer is moving to the left is unknowable yet critical information. Not even if a static, patterned background were present in the five frames, could we confidently state that the ball was moving to the right, because the pattern might have an infinite distance to the observer.

Consider a five-frame clip of a ball moving from the bottom left of a field of vision, to the top right. Motion estimation techniques can determine that on a two dimensional plane the ball is moving up and to the right and vectors describing this motion can be extracted from the sequence of frames. For the purposes of video compression (e.g., [[MPEG]]), the sequence is now described as well as it needs to be. However, in the field of machine vision, the question of whether the ball is moving to the right or if the observer is moving to the left is unknowable yet critical information. Not even if a static, patterned background were present in the five frames, could we confidently state that the ball was moving to the right, because the pattern might have an infinite distance to the observer.

Optical flow has also been applied to [[fluid mechanics]] as a method of estimating flow patterns in a non-invasive way if visible tracer particles are added. This approach is called [[particle image velocimetry]].{{cite journal |last1=Mendes |first1=Luís P. N. |last2=Ricardo |first2=Ana M. C. |last3=Bernardino |first3=Alexandre J. M. |last4=Ferreira |first4=Rui M. L. |title=A comparative study of optical flow methods for fluid mechanics |journal=Experiments in Fluids |date=15 December 2021 |volume=63 |issue=1 |pages=7 |doi=10.1007/s00348-021-03357-7 |url=https://link.springer.com/article/10.1007/s00348-021-03357-7 |language=en |issn=1432-1114|url-access=subscription }}

Optical flow has also been applied to [[fluid mechanics]] as a method of estimating flow patterns in a non-invasive way if visible tracer particles are added. This approach is called [[particle image velocimetry]] (PIV){{cite journal |last1=Mendes |first1=Luís P. N. |last2=Ricardo |first2=Ana M. C. |last3=Bernardino |first3=Alexandre J. M. |last4=Ferreira |first4=Rui M. L. |title=A comparative study of optical flow methods for fluid mechanics |journal=Experiments in Fluids |date=15 December 2021 |volume=63 |issue=1 |pages=7 |doi=10.1007/s00348-021-03357-7 |url=https://link.springer.com/article/10.1007/s00348-021-03357-7 |language=en |issn=1432-1114|url-access=subscription }}. It has been shown that optical flow methods can provide higher accuracy than traditional cross-correlation in PIV processing{{cite journal |last1=Jassal |first1=Gauresh Raj |last2=Schmidt |first2=Bryan E. |date=2025-02-10 |title=A review of optical flow velocimetry in fluid mechanics |journal=Measurement Science and Technology |volume=36 |issue=3 |pages=032002 |doi=10.1088/1361-6501/adafcf |url=https://iop.org}}

.

==Optical flow sensor{{anchor|flow}}==

==Optical flow sensor{{anchor|flow}}==