24/7 Place Recognition by View Synthesis

By Akihiko Torii, Relja Arandjelovi'c, Josef Sivic, Masatoshi Okutomi, Tomás Pajdla. IEEE Transactions on Pattern Analysis and Machine Intelligence, 2018

706 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, Electrical and Electronic Engineering

View Paper: https://www.semanticscholar.org/paper/32c7594171e1882e16b00f07bcc20b6ea9b1b46b

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

In Electrical and Electronic Engineering

Citations in the corpus

Patch-NetVLAD: Multi-Scale Fusion of Locally-Global Descriptors for Place Recognition (2021, methodology): “Global Image Descriptors: Notable early global image descriptor approaches include aggregation of local keypoint descriptors either through a Bag of Words (BoW) scheme [64, 16], Fisher Vectors (FV) [34, 53] or Vector of Locally Aggregated Descriptors (VLAD) [35, 2].”
Visual Camera Re-Localization From RGB and RGB-D Images Using DSAC (2020, methodology): “Given a query image, we search for the most similar data base image by matching global image descriptors, such as DenseVLAD [21], or its learned successor NetVLAD [22].”
Comparative Evaluation of Hand-Crafted and Learned Local Features (2017, background): “, in Structure-fromMotion (SFM) and Multi-View Stereo (MVS) [1,17,33,37, 39,40], image retrieval [31,34,47,48], and image-based localization [35, 36, 56].”
PointNetVLAD: Deep Point Cloud Based Retrieval for Large-Scale Place Recognition (2018, methodology): “Google Street View Time Machine and Tokyo 24/7 [46]).”
Understanding the Limitations of CNN-Based Absolute Camera Pose Regression (2019, methodology): “Standard image retrieval approaches first find the training image J most similar to a given test image I, where similarity is defined in some feature space such as Bag-of-Words (BoW) [67] or VLAD [3,27,71].”
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].”
Back to the Feature: Learning Robust Camera Localization from Pixels to Pose (2021, methodology): “We report image retrieval with DenseVLAD [88] but not PoseNet and its variants as they perform similarly [78].”
InLoc: Indoor Visual Localization with Dense Matching and View Synthesis (2019, methodology): “descriptors [81] or (densely extracted) descriptors based on convolutional neural networks (CNN) [5], [41], [46], [91].”
Learned Contextual Feature Reweighting for Image Geo-Localization (2017, methodology): “Recalls on Tokyo 24/7 [57] and Pittsburgh 250k test [2] datasets.”
NetVLAD: CNN Architecture for Weakly Supervised Place Recognition (2015, methodology): “We follow the standard place recognition evaluation procedure [5, 24, 64, 79, 80].”
From Coarse to Fine: Robust Hierarchical Localization at Large Scale (2018, methodology): “For Active Search [43], City Scale Localization [51], DenseVLAD [56], and NetVLAD [2], we use the evaluation reported in the paper introducing the benchmark.”
Google Landmarks Dataset v2 – A Large-Scale Benchmark for Instance-Level Recognition and Retrieval (2020, background): “7 M images; 24/7 Tokyo [55] containing 1 k images under different illumination conditions and Paris 500 k [60], containing 501 k images.”
D2-Net: A Trainable CNN for Joint Description and Detection of Local Features (2019, methodology): “Several of these methods start by dense descriptor extraction [3,37,60,61] and later aggregate these descriptors into a compact image-level descriptor for retrieval.”
SIFT Meets CNN: A Decade Survey of Instance Retrieval (2016, background): “Examples include place retrieval [167], pedestrian retrieval [168], vehicle retrieval [169], logo retrieval [77], etc.”
Spatial Pyramid-Enhanced NetVLAD With Weighted Triplet Loss for Place Recognition (2020, methodology): “The typical works of this type include [8]–[10], [12]–[14], and [23].”
University-1652: A Multi-view Multi-source Benchmark for Drone-based Geo-localization (2020, background): “Most previous works regard the geo-localization problem as a subtask of image retrieval [2, 16, 18, 21, 31, 32, 37, 39].”