Visual Place Recognition with Repetitive Structures

By Akihiko Torii, Josef Sivic, Masatoshi Okutomi, Tomás Pajdla. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2015

531 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition

View Paper: https://www.semanticscholar.org/paper/c6cf6899d1bb227a80b8ccb1c2836a647e4de1d8

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Aerospace Engineering

In Computer Vision and Pattern Recognition

Citations in the corpus

Image-Based Localization Using LSTMs for Structured Feature Correlation (2016, methodology): “Given a query image, they employ image retrieval techniques to identify the database photo most similar to the query [2,41,50,51,59].”
Efficient Global 2D-3D Matching for Camera Localization in a Large-Scale 3D Map (2017, methodology): “In order to evaluate our method’s resilience to repetitive structure, we adopted (and modified) method of [54] for repeated feature removal.”
Wide-Area Image Geolocalization with Aerial Reference Imagery (2015, background): “The second category of methods match using global image features [11, 15, 45].”
Large-Scale Image Retrieval with Attentive Deep Local Features (2016, background): “Although Pitts250k [35] is larger, it is specialized to visual places with repetitive patterns and may not be appropriate for the general image retrieval task.”
To Aggregate or Not to aggregate: Selective Match Kernels for Image Search (2013, unknown): “structures [28] × 65.”
Learning and Calibrating Per-Location Classifiers for Visual Place Recognition (2013, result): “Similar to other work in large scale place recognition (Cummins and Newman 2009; Knopp et al. 2010; Schindler et al. 2007; Torii et al. 2013) and image retrieval (Nister and Stewenius 2006; Philbin et al.”
Is Context-Aware CNN Ready for the Surroundings? Panoramic Semantic Segmentation in the Wild (2021, methodology): “Examples of automatically generated panoramic annotations for the stitched panoramas using Pittsburgh dataset [63]. for the unlabeled panoramas.”
Image Geo-Localization Based on MultipleNearest Neighbor Feature Matching UsingGeneralized Graphs (2014, background): “Index Terms—Geo-location, image localization, Generalized Minimum Clique Problem (GMCP), generalized minimum spanning tree (GMST), feature matching, multiple nearest neighbor feature matching, feature correspondence, generalized graphs Ç”
Predicting Good Features for Image Geo-Localization Using Per-Bundle VLAD (2015, result): “This setting is similar to those used in [9, 12, 37].”
24/7 Place Recognition by View Synthesis (2018, unknown): “RECENT years have seen a tremendous progress [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16] in the large-scale visual place recognition problem [14], [17].”
Learned Contextual Feature Reweighting for Image Geo-Localization (2017, methodology): “[58], lowers the weights for features occurring in repetitive structure, as they violate feature independence and lead to over counting in bag-of-words representations.”
Spatial Pyramid-Enhanced NetVLAD With Weighted Triplet Loss for Place Recognition (2020, background): “[13] combined the histogram of visual words with the repetitive structure as useful clues for place recognition.”
SIFT Meets CNN: A Decade Survey of Instance Retrieval (2016, background): “When most works try to suppress burstiness, Torii et al [80] view it as a distinguishing feature for architectures and design new similarity measurement following burstiness detection.”
Patch-NetVLAD: Multi-Scale Fusion of Locally-Global Descriptors for Place Recognition (2021, methodology): “Patch-NetVLAD won the Mapillary Challenge at the ECCV2020 workshops (not yet publicly announced to comply with CVPR’s double-blind policy), with Table 1 showing that Patch-NetVLAD outperformed the baseline method, NetVLAD, by 13.0% (absolute R@1 increase) on the withheld test dataset.”
NetVLAD: CNN Architecture for Weakly Supervised Place Recognition (2015, methodology): “We follow the standard place recognition evaluation procedure [5, 24, 64, 79, 80].”
InLoc: Indoor Visual Localization with Dense Matching and View Synthesis (2019, methodology): “[25], [37], [39], [58], [75], [86] and can be further improved by spatial re-ranking [60], informative feature selection [24], [25] or feature weighting [31], [38], [65], [82].”
Geometric Loss Functions for Camera Pose Regression with Deep Learning (2017, methodology): “Traditionally, this has been modelled as an image retrieval problem [6, 9, 53, 45] enabling the use of efficient and scalable retrieval approaches [35, 38] such as Bag-of-Words (BoW) [47], VLAD [17, 10], and Fisher vectors [18].”
Efficient & Effective Prioritized Matching for Large-Scale Image-Based Localization (2017, background): “Over the years, various improvements have been made to handle confusing [15] and repetitive [16] structures, scale to larger scenes [14], [17], or use the original feature descriptors for retrieval [18], [19].”