KRAFT: Keypoint Robust and Adaptive Feature Tracking
TL;DR
KRAFT introduces a robust and adaptive feature tracking method for computer vision, balancing the strengths of learning-based and classical approaches like SIFT and ORB to improve dense correspondence estimation without heavy data or hardware needs.
KRAFT: Keypoint Robust and Adaptive Feature Tracking
Jaroslav Venjarski; Gregor Rozinaj; Ivan Minárik; Šimon Tibenský
https://doi.org/10.1109/ACCESS.2025.3649147
Volume 14
The robust estimation of dense correspondences between images is a cornerstone of computer vision, enabling applications from 3D reconstruction to autonomous navigation. While learning-based methods have recently shown promise, they often require extensive training data and specialized hardware. Conversely, classical methods such as SIFT and ORB, though robust to large viewpoint changes, can be co...