A Pyramid Convolution-Based Scene Coordinate Regression Network for AR-GIS
Bibliographic Data
| ID | 22033733 |
|---|---|
| Authors | Haobo Xu (0000-0002-0243-6516, University of Hong Kong), Chao Zhu (0000-0003-3537-6414, Chongqing Jiaotong University, corresponding author), Yilong Wang (0000-0003-2725-2788, Chongqing Jiaotong University), Huachen Zhu (0000-0003-2711-0501, Chongqing Geomatics and Remote Sensing Center, Chongqing 401120, China), Wei Ma (0000-0002-1727-4206, Chongqing Jiaotong University) |
| Year | 2025 |
| Volume | 14 |
| Issue | 8 |
| Pages | 311 |
| Publication date | 2025-08-15 |
| Peer Reviewed | Yes |
| Open Access | Yes |
| Type | ARTICLE |
| Venue | ISPRS International Journal of Geo-Information (JOURNAL) |
| Journal identifiers | ISSN: 2220-9964 • E-ISSN: 2220-9964 |
| Publisher | MDPI AG (PUBLISHER • IT) |
| DOI | 10.3390/ijgi14080311 |
| OpenAlex | W4413416134 |
| Language | EN |
| References cited | 57 |
Camera tracking plays a pivotal role in augmented reality geographic information systems (AR-GIS) and location-based services (LBS), serving as a crucial component for accurate spatial awareness and navigation. Current learning-based camera tracking techniques, while achieving superior accuracy in pose estimation, often overlook changes in scale. This oversight results in less stable localization performance and challenges in coping with dynamic environments. To address these tasks, we propose a pyramid convolution-based scene coordinate regression network (PSN). Our approach leverages a pyramidal convolutional structure, integrating kernels of varying sizes and depths, alongside grouped convolutions that alleviate computational demands while capturing multi-scale features from the input imagery. Subsequently, the network incorporates a novel randomization strategy, effectively diminishing correlated gradients and markedly bolstering the training process’s efficiency. The culmination lies in a regression layer that maps the 2D pixel coordinates to their corresponding 3D scene coordinates with precision. The experimental outcomes show that our proposed method achieves centimeter-level accuracy in small-scale scenes and decimeter-level accuracy in large-scale scenes after only a few minutes of training. It offers a favorable balance between localization accuracy and efficiency, and effectively supports augmented reality visualization in dynamic environments
Artificial neural network · Computer vision · Geography · Geometry · Regression · Statistics · Advanced Image and Video Retrieval Techniques · Automated Road and Building Extraction · Computer Science · Mathematics · Remote Sensing and LiDAR Applications · Artificial Intelligence
| Citation velocity | historical |
|---|---|
| Highly cited | No |