AutoCompass: Accurate Visual Localization on Public Maps by Learning from Weak Labels

• URL: http://arxiv.org/abs/2609.02798v1
• Date: Wed, 02 Sep 2026 16:31:50 GMT
• Title: AutoCompass: Accurate Visual Localization on Public Maps by Learning from Weak Labels
• Abstract summary: We present AutoSLAM, a supervision approach for training neural map matchers from inaccurate absolute pose labels.<n>We show that heading labels are unnecessary, and that defining a tolerance region around raw GPS improves positional accuracy.<n>Across driving benchmarks, AutoSLAM consistently outperforms counterparts trained with the usual strong reliance on absolute pose labels.
• Score: 31.73754004109431
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: Neural map matchers estimate an image's 3-DoF pose relative to a 2D map. These models are trained on large-scale datasets of geo-referenced images, whose position and heading labels often contain noise that affects the trained models. To address this, we present AutoCompass, a supervision approach for training neural map matchers from inaccurate absolute pose labels. First, we show that heading labels are unnecessary: trained from raw GPS labels, models learn to predict accurate headings, automatically. Second, defining a tolerance region around raw GPS improves positional accuracy. Third, if available, our supervision uses relative poses between training images, obtained via SLAM or SfM, which provide a more accurate training signal. Across driving and egocentric benchmarks, AutoCompass consistently outperforms counterparts trained with the usual strong reliance on absolute pose labels.

Related papers

• PseudoMapTrainer: Learning Online Mapping without HD Maps [41.789167930501016]
  PseudoMapTrainer is a novel approach to online mapping that uses pseudo-labels generated from unlabeled sensor data.<n>We derive those pseudo-labels by reconstructing the road surface from multi-camera imagery.<n>Our pseudo-labels can be effectively used to pre-train an online model in a semi-supervised manner.
  arXiv  Detail & Related papers  (2025-08-26T08:13:30Z)
• GPS as a Control Signal for Image Generation [95.43433150105385]
  We show that the GPS tags contained in photo metadata provide a useful control signal for image generation.<n>We train GPS-to-image models and use them for tasks that require a fine-grained understanding of how images vary within a city.
  arXiv  Detail & Related papers  (2025-01-21T18:59:46Z)
• TrajSSL: Trajectory-Enhanced Semi-Supervised 3D Object Detection [59.498894868956306]
  Pseudo-labeling approaches to semi-supervised learning adopt a teacher-student framework. We leverage pre-trained motion-forecasting models to generate object trajectories on pseudo-labeled data. Our approach improves pseudo-label quality in two distinct manners.
  arXiv  Detail & Related papers  (2024-09-17T05:35:00Z)
• Weakly-supervised Camera Localization by Ground-to-satellite Image Registration [52.54992898069471]
  We propose a weakly supervised learning strategy for ground-to-satellite image registration. It derives positive and negative satellite images for each ground image. We also propose a self-supervision strategy for cross-view image relative rotation estimation.
  arXiv  Detail & Related papers  (2024-09-10T12:57:16Z)
• EgoVM: Achieving Precise Ego-Localization using Lightweight Vectorized Maps [9.450650025266379]
  We present EgoVM, an end-to-end localization network that achieves comparable localization accuracy to prior state-of-the-art methods. We employ a set of learnable semantic embeddings to encode the semantic types of map elements and supervise them with semantic segmentation. We adopt a robust histogram-based pose solver to estimate the optimal pose by searching exhaustively over candidate poses.
  arXiv  Detail & Related papers  (2023-07-18T06:07:25Z)
• CSP: Self-Supervised Contrastive Spatial Pre-Training for Geospatial-Visual Representations [90.50864830038202]
  We present Contrastive Spatial Pre-Training (CSP), a self-supervised learning framework for geo-tagged images. We use a dual-encoder to separately encode the images and their corresponding geo-locations, and use contrastive objectives to learn effective location representations from images. CSP significantly boosts the model performance with 10-34% relative improvement with various labeled training data sampling ratios.
  arXiv  Detail & Related papers  (2023-05-01T23:11:18Z)
• Sparse Semantic Map-Based Monocular Localization in Traffic Scenes Using Learned 2D-3D Point-Line Correspondences [29.419138863851526]
  Given a query image, the goal is to estimate the camera pose corresponding to the prior map. Existing approaches rely heavily on dense point descriptors at the feature level to solve the registration problem. We propose a sparse semantic map-based monocular localization method, which solves 2D-3D registration via a well-designed deep neural network.
  arXiv  Detail & Related papers  (2022-10-10T10:29:07Z)
• 3SD: Self-Supervised Saliency Detection With No Labels [19.260185488168982]
  We present a conceptually simple self-supervised method for saliency detection. Our method generates and uses pseudo-ground truth labels for training.
  arXiv  Detail & Related papers  (2022-03-09T01:40:28Z)
• Semantic Image Alignment for Vehicle Localization [111.59616433224662]
  We present a novel approach to vehicle localization in dense semantic maps using semantic segmentation from a monocular camera. In contrast to existing visual localization approaches, the system does not require additional keypoint features, handcrafted localization landmark extractors or expensive LiDAR sensors.
  arXiv  Detail & Related papers  (2021-10-08T14:40:15Z)
• An Empirical Study of the Collapsing Problem in Semi-Supervised 2D Human Pose Estimation [80.02124918255059]
  Semi-supervised learning aims to boost the accuracy of a model by exploring unlabeled images. We learn two networks to mutually teach each other. The more reliable predictions on easy images in each network are used to teach the other network to learn about the corresponding hard images.
  arXiv  Detail & Related papers  (2020-11-25T03:29:52Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.