Optimal Reduction of Large Image Databases for Location Recognition

Havlena, Michal; Hartmann, Wilfried; Schindler, Konrad

doi:10.1109/iccvw.2013.93

Cited by 12 publications

(5 citation statements)

References 30 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The proposed method is validated using two datasets sourced from Flickr [23]. The first one, MERGED, was kindly provided by the authors of [5]. It consists of 9,469 images from three different landmarks: roughly one third of them depicts the Fontana di Trevi in Rome, another third the Duomo in Milan, and the last third the Old Town Square in Prague.…”

Section: Methodsmentioning

confidence: 99%

VocMatch: Efficient Multiview Correspondence for Structure from Motion

Havlena

Schindler

2014

Computer Vision – ECCV 2014

Self Cite

View full text Add to dashboard Cite

Feature matching between pairs of images is a main bottleneck of structure-from-motion computation from large, unordered image sets. We propose an efficient way to establish point correspondences between all pairs of images in a dataset, without having to test each individual pair. The principal message of this paper is that, given a sufficiently large visual vocabulary, feature matching can be cast as image indexing, subject to the additional constraints that index words must be rare in the database and unique in each image. We demonstrate that the proposed matching method, in conjunction with a standard inverted file, is 2-3 orders of magnitude faster than conventional pairwise matching. The proposed vocabulary-based matching has been integrated into a standard SfM pipeline, and delivers results similar to those of the conventional method in much less time.

show abstract

Section: Methodsmentioning

confidence: 99%

VocMatch: Efficient Multiview Correspondence for Structure from Motion

Havlena

Schindler

2014

Computer Vision – ECCV 2014

Self Cite

View full text Add to dashboard Cite

show abstract

“…While heuristics for the selection as used by Li et al (2010) and Dymczyk et al (2015b) provide good results, it has been shown by Mauro et al (2014) and Havlena et al (2013) that optimization-based approaches can provide gains especially under high compression rates. These approaches target picking the most informative landmarks while ensuring that every camera must observe at least N thres landmarks to provide coverage across the model area.…”

Section: The Need For Model Compressionmentioning

confidence: 99%

Large-scale, real-time visual–inertial localization revisited

Lynen

Zeisl

Aiger

et al. 2020

The International Journal of Robotics Research

View full text Add to dashboard Cite

The overarching goals in image-based localization are scale, robustness, and speed. In recent years, approaches based on local features and sparse 3D point-cloud models have both dominated the benchmarks and seen successful real-world deployment. They enable applications ranging from robot navigation, autonomous driving, virtual and augmented reality to device geo-localization. Recently, end-to-end learned localization approaches have been proposed which show promising results on small-scale datasets. However, the positioning accuracy, scalability, latency, and compute and storage requirements of these approaches remain open challenges. We aim to deploy localization at a global scale where one thus relies on methods using local features and sparse 3D models. Our approach spans from offline model building to real-time client-side pose fusion. The system compresses the appearance and geometry of the scene for efficient model storage and lookup leading to scalability beyond what has been demonstrated previously. It allows for low-latency localization queries and efficient fusion to be run in real-time on mobile platforms by combining server-side localization with real-time visual–inertial-based camera pose tracking. In order to further improve efficiency, we leverage a combination of priors, nearest-neighbor search, geometric match culling, and a cascaded pose candidate refinement step. This combination outperforms previous approaches when working with large-scale models and allows deployment at unprecedented scale. We demonstrate the effectiveness of our approach on a proof-of-concept system localizing 2.5 million images against models from four cities in different regions of the world achieving query latencies in the 200 ms range.

show abstract

“…One popular visual map compression method is based on optimization approach, formulating the selection problem as a programming problem. For computer vision tasks, location recognition is similar with the localization, and ILP was used in data database to select the most compact subset of images to achieve recognition [13]. Park et al [8] proposed the quadratic programming (QP) method to solve the visual map compression, which is a baseline for further 3D point cloud reduction.…”

Section: Related Workmentioning

confidence: 99%

3D LiDAR Map Compression for Efficient Localization on Resource Constrained Vehicles

Yin

Wang

Tang

et al. 2021

IEEE Trans. Intell. Transport. Syst.

View full text Add to dashboard Cite

Large scale 3D maps constructed via LiDAR sensor are widely used on intelligent vehicles for localization in outdoor scenes. However, loading, communication and processing of the original dense maps are time consuming for onboard computing platform, which calls for a more concise representation of maps to reduce the complexity but keep the performance of localization. In this paper, we propose a teacher-student learning paradigm to compress the 3D point cloud map. Specifically, we first find a subset of LiDAR points with high number of observations to preserve the localization performance, which is regarded as the teacher of map compression. An efficient optimization strategy is proposed to deal with the massive data in original map. With the supervision of compressed map, a student model is built by training a random forest model fed with geometric feature descriptors of each point. As a result, the student model is able to compress the map without referring to the expensive numerical optimization. Additionally, by incorporating the features, the innovative student model can be generalized to other new maps while no re-training is required. We conduct thorough experiments on multi-session dataset and KITTI dataset to demonstrate the effectiveness and efficiency of the proposed learning paradigm, and the comparison with other map compression methods. The final results show that the learned student model can achieve efficient map compression with comparable LiDAR based localization performance to the original map at the same time.

show abstract

Optimal Reduction of Large Image Databases for Location Recognition

Cited by 12 publications

References 30 publications

VocMatch: Efficient Multiview Correspondence for Structure from Motion

VocMatch: Efficient Multiview Correspondence for Structure from Motion

Large-scale, real-time visual–inertial localization revisited

3D LiDAR Map Compression for Efficient Localization on Resource Constrained Vehicles

Contact Info

Product

Resources

About