Elsevier

Real-Time Imaging

Volume 3, Issue 2, April 1997, Pages 71-86
Real-Time Imaging

Regular Article
Real-Time Quantized Optical Flow

https://doi.org/10.1006/rtim.1996.0048Get rights and content

Abstract

Algorithms based on the correlation of image patches can be robust in practice but are computationally intensive due to the computational complexity of their search-based nature. Performing the search over time instead of over space is linear in nature, rather than quadratic, and results in a very efficient algorithm. This, combined with implementations which are highly efficient on standard computing hardware, yields performance of 9 frame/sec on a scientific workstation. Although the resulting velocities are quantized with resulting quantization error, they have been shown to be sufficiently accurate for many robotic vision tasks such as time-to-collision and robotic navigation. Thus, this algorithm is highly suitable for real-time robotic vision research.

References (0)

Cited by (50)

  • Optical flow-based representation for video action detection

    2015, Emerging Trends in Image Processing, Computer Vision and Pattern Recognition
  • Autonomous quadrotor flight with vision-based obstacle avoidance in virtual environment

    2012, Expert Systems with Applications
    Citation Excerpt :

    In frequency based techniques, spatiotemporal velocity-tuned linear filters are utilized to create the new form of the image sequence, and optical flow velocity matrix is obtained from the new form of the image sequence (Fleet & Jepson, 1990). In the correlation based method, features are extracted from sequential images and optical flow is calculated as a matching feature obtained using the consecutive images (Camus, 1997). Differential optical flow techniques take the advantage of spatiotemporal derivatives of image sequences (Barron et al., 1994).

View all citing articles on Scopus

This research was conducted while the author held a National Research Council Research Associateship at NIST.

f2

E-mail: [email protected]. URL: http://isd.cme.nist.gov/staff/camus/

View full text