A Robust Pixel-Aware Gyro-Aided KLT Feature Tracker for Large Camera Motions

Weibo Huang,Hong Liu

DOI: https://doi.org/10.1109/tim.2021.3129493

IF: 5.6

2021-01-01

IEEE Transactions on Instrumentation and Measurement

Abstract:Tracking fails in the optimization step of conventional Kanade-Lucas–Tomasi (KLT) feature tracker mainly due to the inadequate initial condition that falls out of the convergence region, especially when a camera rotates rapidly or shakes severely. To overcome the problem, we propose a pixel-aware gyro-aided KLT feature tracker that remains accurate and robust under fast camera-ego motion conditions. In particular, we develop a pixel-aware gyro-aided feature prediction algorithm to predict the initial optical flow and obtain the patch deformation matrix of each feature point. It increases the probability of initial estimates to locate in its convergence region. Unlike the existing methods, which assume all the tracked feature pairs were constrained by the same homography prediction matrix, our prediction matrix is adjustable for each feature as it considers the pixel coordinates in the prediction process. A geometric validation based on homography and fundamental check is also adopted to remove outlier tracks. Experimental results on both public datasets and real-world sequences demonstrate that the feature tracking accuracy and robustness can be significantly improved by the proposed method. To facilitate further development, the code is publicly available at https://github.com/weibohuang0314/pixel_aware_gyro_aided_klt_feature_tracker.