|TREND||DATASET||BEST METHOD||PAPER TITLE||PAPER||CODE||COMPARE|
Maps are a key component in image-based camera localization and visual SLAM systems: they are used to establish geometric constraints between images, correct drift in relative pose estimation, and relocalize cameras after lost tracking.
The most promising approach is inspired by reinforcement learning, namely to replace the deterministic hypothesis selection by a probabilistic selection for which we can derive the expected loss w. r. t.
Popular research areas like autonomous driving and augmented reality have renewed the interest in image-based camera localization.
In contrast, we learn hypothesis search in a principled fashion that lets us optimize an arbitrary task loss during training, leading to large improvements on classic computer vision tasks.
We thus propose to jointly learn keypoint detection and description together with a predictor of the local descriptor discriminativeness.
In this work, we fit the 6D camera pose to a set of noisy correspondences between the 2D input image and a known 3D environment.
Additionally, the dropout module enables the pose regressor to output multiple hypotheses from which the uncertainty of pose estimates can be quantified and leveraged in the following uncertainty-aware pose-graph optimization to improve the robustness further.