Augmented Reality Pose Estimation: A Systematic Review of Visual Localization and SLAM
Abstract
Augmented Reality (AR) has attracted increasing attention as it enhances user perception and interaction by overlaying virtual content onto the physical world. Accurate and real-time six degrees of freedom (6-DoF) pose estimation is a core requirement for reliable spatial registration. To help researchers understand recent advances and select appropriate methods for AR applications, this survey provides a systematic overview of AR pose estimation algorithms from two complementary perspectives: visual localization with prior scene knowledge and simultaneous localization and mapping (SLAM), which performs online pose estimation and mapping in unknow environments. For visual localization, pose estimation methods can be grouped into three major lines, including feature-matching-based localization, scene coordinate regression, and pose regression; their evolution toward improved robustness and scalability is also discussed. For SLAM-based pose estimation, representative approaches are organized into traditional visual SLAM (VSLAM), deep learning–enhanced SLAM, rendering-based SLAM, highlighting key design choices as well as the strengths and limitations of each category under AR constraints. In addition, evaluation metrics and commonly used benchmarks are reviewed, and reported performance of representative algorithms on selected datasets is summarized. Finally, the review summarizes the current state of AR pose estimation and outlines future research directions.
How strong is the evidence?
Evidence Hub reads this as a systematic review. Its evidence level is 1 of 5 (Highest · pooled evidence). Pooled evidence from several studies sits at the top of the evidence pyramid.
Read the paper
Related studies
- Knowledge Management Perspective of Generative Artificial Intelligence
- Navigating Generative Artificial Intelligence Promises and Perils for Knowledge and Creative Work
- Characterizing the evolving research on enterprise content management
- Large language models and multimodal foundation models for precision oncology
- From Generative AI to Extended Reality: Multidisciplinary Perspectives on the Challenges, Opportunities, and Future of Educational Computing
- Generative Artificial Intelligence: A New Engine for Advancing Environmental Science and Engineering
Go deeper in Evidence Hub
Evidence Hub adds a study data sheet (design, sample size, evidence level, key numbers), an AI summary and a chat grounded in the paper's text.