Automatic differentiation facilitates OF-integration into steering-angle-based road vehicle tracking

摘要: The interpretation of road traffic scenes recorded by a stationary video camera requires sufficiently large field view in order to capture non-trivial maneuvers. Vehicle images will thus be small. Their reliable tracking needs combine edge-based approaches for precision with area-oriented approaches, example based on match between estimated and predicted Optic Flow fields, stability. approach reported here estimates the steering angle vehicles tracked compares this as well velocity estimate actual measurements performed suitably equipped test vehicle while crossing intersection. In addition, results numerous other same scene have been assessed interactively are reported. Computation Jacobi matrix Kalman-Filter is facilitated "dual number" representation partial derivatives.