Automatic Relocalization and Loop Closing for Real-Time Monocular SLAM

Abstract

Monocular SLAM has the potential to turn inexpensive cameras into powerful pose sensors for applications such as robotics and augmented reality. We present a relocalization module for such systems which solves some of the problems encountered by previous monocular SLAM systems--tracking failure, map merging, and loop closure detection. This module extends recent advances in keypoint recognition to determine the camera pose relative to the landmarks within a single frame time of 33 ms. We first show how this module can be used to improve the robustness of these systems. Blur, sudden motion, and occlusion can all cause tracking to fail, leading to a corrupted map. Using the relocalization module, the system can automatically detect and recover from tracking failure while preserving map integrity. Extensive tests show that the system can then reliably generate maps for long sequences even in the presence of frequent tracking failure. We then show that the relocalization module can be used to recognize overlap in maps, i.e., when the camera has returned to a previously mapped area. Having established an overlap, we determine the relative pose of the maps using trajectory alignment so that independent maps can be merged and loop closure events can be recognized. The system combining all of these abilities is able to map larger environments and for significantly longer periods than previous systems.

Fast relocalisation and loop closing in keyframe-based SLAM

“…Williams et al [10] perform relocalisation and loop closing on a SLAM system, based on a filter approach. They use randomized list of binary test classifiers to find correspondences between image features and previously trained map features.…”

Real-time RGB-D camera relocalization

“…The first category are landmark-based approaches (LbAs) [3,26]. During successful tracking, fiducial landmarks, also called keypoints, are extracted from the camera images, encoded by a descriptor, and stored in a database together with their 3D locations.…”

“…Also, the construction of the keypoint database in real-time settings can be challenging. Often a costly online training phase is required which demands additional resources such as a background thread or extra GPU computations [26]. Another limitation of LbAs lies in their inherent sparse representation of the scene.…”

“…Our method makes use of randomized ferns which have been previously used in the context of keypoint-based relocalization [26]. The way in which we make use of ferns is quite different and will be described in Section 2.…”

Matching-range-constrained real-time loop closure detection with CNNs features

“…The state-of-the-art algorithms take advantage of the bag-of-words (BoW) model [1][2][3] to describe images. The BoW model clusters the visual feature descriptors in images, builds the dictionary, and then finds the corresponding words of each image.…”