USACv20: robust essential, fundamental and homography matrix estimation
This provides incremental improvements for computer vision researchers and practitioners needing robust geometric model estimation.
The authors combined the best recent RANSAC-like robust estimators to create USACv20, a modular framework for estimating homographies, fundamental and essential matrices. On eight real-world datasets, it achieved the most geometrically accurate models and fastest speed compared to state-of-the-art methods.
We review the most recent RANSAC-like hypothesize-and-verify robust estimators. The best performing ones are combined to create a state-of-the-art version of the Universal Sample Consensus (USAC) algorithm. A recent objective is to implement a modular and optimized framework, making future RANSAC modules easy to be included. The proposed method, USACv20, is tested on eight publicly available real-world datasets, estimating homographies, fundamental and essential matrices. On average, USACv20 leads to the most geometrically accurate models and it is the fastest in comparison to the state-of-the-art robust estimators. All reported properties improved performance of original USAC algorithm significantly. The pipeline will be made available after publication.