Scene Coordinate Regression with Angle-Based Reprojection Loss for Camera Relocalization
This work addresses camera relocalization for computer vision and robotics applications, presenting an incremental improvement over existing methods.
The paper tackles the problem of camera relocalization by introducing an angle-based reprojection loss that eliminates the need for careful network initialization and improves accuracy, achieving more accurate results as demonstrated by performance gains.
Image-based camera relocalization is an important problem in computer vision and robotics. Recent works utilize convolutional neural networks (CNNs) to regress for pixels in a query image their corresponding 3D world coordinates in the scene. The final pose is then solved via a RANSAC-based optimization scheme using the predicted coordinates. Usually, the CNN is trained with ground truth scene coordinates, but it has also been shown that the network can discover 3D scene geometry automatically by minimizing single-view reprojection loss. However, due to the deficiencies of the reprojection loss, the network needs to be carefully initialized. In this paper, we present a new angle-based reprojection loss, which resolves the issues of the original reprojection loss. With this new loss function, the network can be trained without careful initialization, and the system achieves more accurate results. The new loss also enables us to utilize available multi-view constraints, which further improve performance.