SKD: Keypoint Detection for Point Clouds using Saliency Estimation
This work addresses the need for more effective keypoint detection in 3D computer vision, offering significant gains for applications such as autonomous driving and robotics, though it is incremental as it builds on existing deep learning descriptors.
The paper tackles the problem of keypoint detection in point clouds for tasks like registration and reconstruction by introducing SKD, a method that uses saliency estimation based on descriptor gradients, achieving up to 50% improvement in matchability and repeatability on LIDAR datasets.
We present SKD, a novel keypoint detector that uses saliency to determine the best candidates from a point cloud for tasks such as registration and reconstruction. The approach can be applied to any differentiable deep learning descriptor by using the gradients of that descriptor with respect to the 3D position of the input points as a measure of their saliency. The saliency is combined with the original descriptor and context information in a neural network, which is trained to learn robust keypoint candidates. The key intuition behind this approach is that keypoints are not extracted solely as a result of the geometry surrounding a point, but also take into account the descriptor's response. The approach was evaluated on two large LIDAR datasets - the Oxford RobotCar dataset and the KITTI dataset, where we obtain up to 50% improvement over the state-of-the-art in both matchability and repeatability. When performing sparse matching with the keypoints computed by our method we achieve a higher inlier ratio and faster convergence.