Animals on the Web

By Tamara L. Berg, D.A. Forsyth. 2006

215 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

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

Citations in the corpus

Webly Supervised Learning of Convolutional Networks (2015, background): “But more data needs more human labeling efforts.”
A Multi-View Embedding Space for Modeling Internet Images, Tags, and Their Semantics (2012, background): “A task related to tag-to-image search, though one we do not consider directly, is re-ranking of contaminated image search results for the purpose of dataset collection (Berg and Forsyth, 2006; Fan et al., 2010; Frankel et al., 1997; Schroff et al., 2007).”
LabelMe: A Database and Web-Based Tool for Image Annotation (2008, background): “It has been shown that imag e fe tures can be used to improve the quality of the ranking returned by online queries [14, 3] .”
Joint Video and Text Parsing for Understanding Events and Answering Queries (2013, methodology): “…et al. [2] and Blei [16] discussed a series of probabilistic models to capture the dependencies between image elements and words; Berg and Forsyth [17] demonstrated a method for identifying images containing categories of animals based on word and visual cues; Monay and Gatica-Perez [3] applied…”
Deep Convolutional Ranking for Multilabel Image Annotation (2013, background): “Modeling Internet images and their corresponding textural information (e.g., sentences, tags) have been of great interest in the vision community [2, 10, 12, 15, 30, 29, 34].”
Exploiting weakly-labeled Web images to improve object classification: a domain adaptation approach (2010, methodology): “Subsequent approaches [2, 25, 20] have employed similar outlier reject ion schemes to automatically construct clean(er) data sets of images for training and testing objec t classifiers.”
Learning to recognize objects in egocentric activities (2011, background): “Recent works have tried to provide different sources of automatic weakly supervised annotations by using web data [4] or cheap human annotation systems such as Amazon’s Mechanical Turk.”
Learning Object Categories From Internet Image Searches (2010, background): “Berg and Forsyth [3] overcome the limitations of Google's image search by using text on the original web pages to extract further contextual cues.”
Unsupervised Learning of Visual Sense Models for Polysemous Words (2008, result): “Category classification experiments show that our dictionarybased approach outperforms baseline methods.”
High level describable attributes for predicting aesthetics and interestingness (2011, background): “For “presence of animals” this is the Animals on the Web dataset [2] with images from all 10 animal categories merged into an animal superclass.”
Computing iconic summaries of general visual concepts (2008, background): “Several recent approaches have considered the problem of re-ranking Internet image search results for the purpose of automatically collecting large image databases [3, 9, 16, 22].”
Towards total scene understanding: Classification, annotation and segmentation in an automatic framework (2009, methodology): “Finally, our approach is inspired by earlier approaches of learning from noisy internet data such as [2, 19, 10, 11, 29].”
Contextualizing Object Detection and Classification (2015, background): “Index Terms— Object classiﬁcation, object detection, context modeling Ç 1 I NTRODUCTION R ECOGNIZING objects in an image requires combining many different signals from the raw image data.”
Multiple-instance ranking: Learning to rank images for image retrieval (2008, background): “And there has been much fascination with searching animals on the web recently [3].”
Semantic Label Sharing for Learning with Many Categories (2010, background): “These include: Li et al. [8], Fergus et al. [9], Berg et al. [10], amongst others.”
Keywords to visual categories: Multiple-instance learning forweakly supervised object categorization (2008, methodology): “In the future we are interested in exploring complementary text features within this framework, and considering how prior knowledge about a category’s expected sparsity might be captured in order to boost accuracy.”
Baby talk: Understanding and generating simple image descriptions (2011, background): “Other object classes that have been considered include animal images from the web [7], [30], [37] where text from the containing webpage can be utilized for improved image classification.”