Object retrieval with large vocabularies and fast spatial matching

By James Philbin, Ondřej Chum, Michael Isard, Josef Sivic, Andrew Zisserman. 2007

3,312 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Information Retrieval & Search

How researchers see this paper

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

In Computer Vision and Pattern Recognition

In Information Retrieval & Search

Citations in the corpus

Aggregating Local Image Descriptors into Compact Codes (2012, methodology): “This corresponds to 4 times the recall@4 measure, i.e. , the best performance is 4.”
CPlaNet: Enhancing Image Geolocalization by Combinatorial Partitioning of Maps (2018, background): “While global features have the beneﬁt to retrieve diverse natural scene images based on ambient information, local image features yield higher precision in retrieving structured objects such as buildings and are thus more frequently used [8,9,10,11,12,13,14,15,16].”
Revisiting IM2GPS in the Deep Learning Era (2017, background): “…scene retrieval [33, 2, 23, 7]; Since this line of work mostly focuses on instance-level matching, benchmarks designed for this task consist of popular scenes or landmarks [21, 22, 12] and similarity between matched images are visually recognizable (by humans or geometric veriﬁcation).”
PlaNet - Photo Geolocation with Convolutional Neural Networks (2016, background): “Most local feature based approaches therefore focus on localization within cities, either based on photos from photo sharing websites [5, 36] or street view [2, 7, 26, 27, 43, 51, 52].”
Grafit: Learning fine-grained image representations with coarse labels (2020, methodology): “For category-level retrieval we report the mean average precision (mAP), as commonly done for retrieval tasks [2, 42].”
Soft-NMS — Improving Object Detection with One Line of Code (2017, methodology): “It was first employed in edge detec ion techniques [22].”
CNN Features Off-the-Shelf: An Astounding Baseline for Recognition (2014, methodology): “We report retrieval results on five common datasets in the area as follows: Oxford5k buildings[34] This is a collection of 5063 reference photos gathered from flickr, and 55 queries of different buildings.”
Deep Clustering for Unsupervised Learning of Visual Features (2018, neutral): “, Oxford Buildings [48] and Paris [49].”
Fast ConvNets Using Group-Wise Brain Damage (2015, background): “780 mAP, and a reasonably small drop for the Oxford Building dataset [39] from 0.”
NetVLAD: CNN Architecture for Weakly Supervised Place Recognition (2015, methodology): “9% on Oxford 5k [53], Paris 6k [54], Holidays [26], respectively; for example, this is a +20% relative improvement on Oxford 5k.”
Particular object retrieval with integral max-pooling of CNN activations (2015, methodology): “Geometric matching models (Philbin et al., 2007; Avrithis & Tolias, 2014) are typically applied in a pairwise manner during the re-ranking stage of a short-list of images.”
From generic to specific deep representations for visual recognition (2014, background): “The wide range of visual recognition tasks we consider are: 1) Object image classification (pascal VOC 2007 [9]), 2) Scene image classification (MIT indoor scenes 67[26], SUN 397 scenes [32]), 3) Object attribute detection (H3D human attributes [6], UIUC 64 object attributes [10]), 4) Fine grained c”
Unidentified Video Objects: A Benchmark for Dense, Open-World Segmentation (2021, background): “Many real-world applications such as object search [27, 30], instance registration [50], human-object interaction modeling [14] and human activity understanding [6] require such open-world prediction abilities, e.”
End-to-End Learning of Deep Visual Representations for Image Retrieval (2016, methodology): “An advantage of these techniques is that spatial verification can be employed to rerank a shortlist of results (Philbin et al. 2007; Perdoch et al. 2009), yielding a large improvement despite a significant cost.”
Handling Urban Location Recognition as a 2D Homothetic Problem (2010, background): “It can therefore be diﬃcult to ﬁnd strong geometrical constraints for ﬁltering the initial visual-word based results, although recent approaches look at (locally) consistent orientations and feature shapes [16,17,18] and exploit that pictures are usually not taken upside down.”
Face Recognition and Retrieval Using Cross-Age Reference Coding With Cross-Age Celebrity Dataset (2015, methodology): “been widely used as an evaluation metric in information retrieval and image retrieval tasks [38], and it is calculated as”
Objectness to improve the bag of visual words model (2014, methodology): “The vector quantization process is done by k-means clustering [17], and visual words are stored in a randomized forest of k-d trees [6] to reduce search cost.”