Landmark classification in large-scale image collections

By Yunpeng Li, David J. Crandall, Dan Huttenlocher. 2009

306 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Media Technology

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 Media Technology

Citations in the corpus

CPlaNet: Enhancing Image Geolocalization by Combinatorial Partitioning of Maps (2018, methodology): “The database serves as the index of an image retrieval system [28,29,30,31,32,33] or the training data of a landmark classifier [34,35,36].”
Revisiting IM2GPS in the Deep Learning Era (2017, background): “One natural approach to the image geolocalization task would be to to treat it like an instance retrieval task and match local features from the query image (and perhaps their geometric layout) to a reference database of images with known locations [16].”
PlaNet - Photo Geolocation with Convolutional Neural Networks (2016, methodology): “While [6] train a Hidden Markov Model (HMM) on a dataset of photo albums to learn popular tourist routes, [29] train a structured SVM that uses temporal information as an additional feature.”
Improving the Fisher Kernel for Large-Scale Image Classification (2010, unknown): “This is in contrast with linear SVMs whose training cost is in O(N) [11,12] and which can therefore be efficiently learned with large quantities of images [13].”
Avoiding Confusing Features in Place Recognition (2010, background): “[6] discover distinct but coarse-level landmarks (such as an entire city square) as places with high concentration of geotagged Flickr images and build image-level classifiers to distinguish landmarks from each other.”
Geolocation Estimation of Photos Using a Hierarchical Model and Scene Classification (2018, background): “Approaches that focus on landmark recognition use either a pre-defined set of landmarks or cluster a given photo collection in an unsupervised manner to retrieve the most interesting areas for geolocalization [4,23,28,48].”
Worldwide Pose Estimation Using 3D Point Clouds (2012, unknown): “, the “landmark recognition/classification” problem [12, 13].”
Image Labeling on a Network: Using Social-Network Metadata for Image Classification (2012, methodology): “Another commonly used source of metadata comes directly from the camera, in the form of exif and GPS data [16, 14, 12, 11].”
Computer Vision for Autonomous Vehicles: Problems, Datasets and State-of-the-Art (2017, background): “Early image-based techniques (Li et al. (2009); Zheng et al. (2009)) approach the problem as classification into one of a predefined set of places which are referred as “landmarks”.”
Graph-Based Discriminative Learning for Location Recognition (2013, background): “However, there is no single definition for what it means to be a location, and, accordingly, a wide variety of representations for places have been used in research: Are places, for instance, a set of distinct landmarks, each represented by a set of images? [33, 16] Are places latitude and longitude”
Visual landmark recognition from Internet photo collections: A large-scale evaluation (2014, methodology): “[27] learn a multi-class SVM and additionally use the BoWs (bags-of-words) of the textual tags of the images as features.”
Leveraging Structure from Motion to Learn Discriminative Codebooks for Scalable Landmark Classification (2013, methodology): “Large-scale and finegrained classification has recently gained interest amongst researchers, however, with the exception of [16, 8], we are not aware of much work focusing on landmark or location classification within this framework.”
Multimodal semi-supervised learning for image classification (2010, methodology): “These results are in line with those of [13], where visual features and tags were combined for landmark classification.”
Wide-Area Image Geolocalization with Aerial Reference Imagery (2015, background): “In recent years, a plethora of methods for automatic image geolocalization have been introduced [1, 8, 11, 19, 23, 45].”
Predicting Geo-informative Attributes in Large-Scale Image Collections Using Convolutional Neural Networks (2015, methodology): “These techniques work well for popular landmarks, but cannot infer geo-tags for photos taken outside these places.”
AutoCaption: Automatic caption generation for personal photos (2014, methodology): “We treat each pre-determined landmark as a different class [16].”
Web-Scale Image Clustering Revisited (2015, background): “With few exceptions [3], most methods focus on landmarks and points of interest [11, 23].”
Retrieving landmark and non-landmark images from community photo collections (2010, methodology): “For instance, [6] and [18] use mean-shift to perform geo-clustering alone and mine high-density locations corresponding to popular places.”
High-dimensional signature compression for large-scale image classification (2011, background): “[16] study landmark classification on a collection of 500 landmarks and 2 million images.”