Automatic image annotation and retrieval using cross-media relevance models

By Jiwoon Jeon, Victor Lavrenko, Raghavan Manmatha. Annual International ACM SIGIR Conference on Research and Development in Information Retrieval, 2003

1,351 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Information Systems

View Paper: https://www.semanticscholar.org/paper/228029e7533e32a025071e31e3f4f08d2bea5f5a

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 Systems

Citations in the corpus

Up next: retrieval methods for large scale related video suggestion (2014, methodology): “There is a large variety of methods described in the scien-tiﬁc literature (e.g., [16, 27, 28, 29, 30] among many others) on how to derive a set of annotations for images and videos.”
Artificial intelligence in the creative industries: a review (2020, methodology): “Analysis Ads/film analysis [65] [65] [65] GP [66] Content retrieval [67–70] [71] [72] PM [73] Fake detection [74,75] [74] Blockchain [76] Recommendation [77] Deep belief net [78] Contrast enhancement [79] [80] Histogram [81].”
Modeling annotated data (2003, background): “A number of recent papers have considered generative probabilistic models for such multi-type or relational data [2, 6, 4, 13].”
Automatic multimedia cross-modal correlation discovery (2004, methodology): “…to associate words with images for automatic image captioning , people have proposed meth-ods based on classi(cid:12)ers [15] or information retrieval techniques (relevance model [10] and user feedback [26]), as well as building association models (translation model [7]; hierarchical model [2, 3,…”
Learning Cross-Media Joint Representation With Sparse and Semisupervised Regularization (2014, methodology): “Although some methods are proposed to model image data complemented with text annotations [8], [9], they are restricted to exploiting keywords associated to visible objects.”
OPTIMOL: Automatic Online Picture Collection via Incremental Model Learning (2007, methodology): “One major group of research (Barnard and Forsyth 2001; Barnard et al. 2003; Jeon et al. 2003) in CBIR treats images as a collection of blobs or blocks, each corresponding to a word or phrase in the caption (with some considerable variations).”
Simultaneous image classification and annotation (2009, neutral): “For image annotation, several studies have explored the use of probabilistic models to learn the relationships between images and annotation terms [1, 8, 12].”
Baselines for Image Annotation (2010, methodology): “…has been a topic of ongoing research for more than a decade leading to several interesting techniques (Duygulu et al. 2002; Blei et al. 2003b; Jeon et al. 2003; Wang et al. 2004; Lavrenko et al. 2004; Monay and Gatica-Perez 2003; Feng et al. 2004; Barnard and Johnson 2005; Metzler and…”
On the Role of Correlation and Abstraction in Cross-Modal Multimedia Retrieval (2014, background): “For example, in [29] a query text, and in [30] a query image is used to retrieve similar text documents and images, based on low-level text (e.g., words) and image (e.g., DCTs) representations, respectively.”
Efficient Incremental Learning for Mobile Object Detection (2019, methodology): “…cloud server or locally on a mobile GPU for a fully on-device system, and both have been implemented for prototype system evaluation. problem is automatic image annotation [17, 4], which is a process that relies on the computer system assigning metadata (e.g., keywords, caption) to digital images.”
Heterogeneous Metric Learning with Joint Graph Regularization for Cross-Media Retrieval (2013, background): “Many research efforts have been devoted to the content-based multimedia retrieval (Jeon, Lavrenko, and Manmatha 2003; Greenspan, Goldberger, and Mayer 2004; Escalante et al. 2008).”
A review on automatic image annotation techniques (2012, methodology): “A typical MIML is achieved using probabilistic tools such as the Bayesian methods [18,20,56,60,65].”
Variational inference for Dirichlet process mixtures (2006, methodology): “6 Image analysis Finite Gaussian mixture models are widely used in computer vision to model natural images for the purposes of automatic clustering, retrieval, and classification (Barnard et al. 2003; Jeon et al. 2003).”
Finding meaning on YouTube: Tag recommendation and category discovery (2010, methodology): “In the case of generative models, the value of the joint probability of visual features and labels are used to compute the scores [9].”
Automatic Caption Generation for News Images (2013, methodology): “This task, on its own, is of significant importance for many image-based applications, such as image retrieval, picture browsing support, and story picturing (Lavrenko et al., 2003; Jeon et al., 2003; Blei and Jordan, 2003; Joshi et al., 2006; Li and Wang, 2006).”