FM-Net: Deep Learning Network for the Fundamental Matrix Estimation from Biplanar Radiographs

Abstract

Background and objective: The fundamental matrix estimation is a classic problem in computer vision. The traditional algorithms require high-precision correspondences. However, correspondences in biplanar radiographs are difficult to match accurately.

Methods: We propose an end-to-end network to estimate the F-Matrix directly from BR, which includes feature extraction and regression prediction. There is no publicly available dataset of biplanar radiographs. We produce the dataset in this paper to train and test the proposed network. Four metrics, Mean Square Error, Calculating R-squared, Square Value of Extreme Constraint, and Absolute Value of Extreme Constraint are used to measure the performance of the approaches.

Results: The best Square Value of Extreme Constraint and Absolute Value of Extreme Constraint values we obtained on the datasets were 0.20 and 0.43, respectively. Compared with other methods, the estimation accuracy of FM-Net is improved by more than 53.53%.

Conclusions: The results of experiments demonstrate that the proposed network can estimate the fundamental matrix successfully. It outperforms the classical algorithms and other deep learning-based methods.

Keywords: Biplanar Radiographic Images; Feature Extraction; Fundamental Matrix; Regression Prediction.