Add to Chrome
Add to Firefox
Add to EdgeOnline Photometric Calibration for Auto Exposure Video for Realtime Visual Odometry and SLAM
Oct 05, 2017
Recent direct visual odometry and SLAM algorithms have demonstrated impressive levels of precision. However, they require a photometric camera calibration in order to achieve competitive results. Hence, the respective algorithm cannot be directly applied to an off-the-shelf-camera or to a video sequence acquired with an unknown camera. In this work we propose a method for online photometric calibration which enables to process auto exposure videos with visual odometry precisions that are on par with those of photometrically calibrated videos. Our algorithm recovers the exposure times of consecutive frames, the camera response function, and the attenuation factors of the sensor irradiance due to vignetting. Gain robust KLT feature tracks are used to obtain scene point correspondences as input to a nonlinear optimization framework. We show that our approach can reliably calibrate arbitrary video sequences by evaluating it on datasets for which full photometric ground truth is available. We further show that our calibration can improve the performance of a state-of-the-art direct visual odometry method that works solely on pixel intensities, calibrating for photometric parameters in an online fashion in realtime.
