A New Baseline for Image Annotation

By Ameesh Makadia, Vladimir I. Pavlovic, Sanjiv Kumar. 2008

511 citations recorded by the source corpus.

Subfields: Artificial Intelligence, 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 Artificial Intelligence

In Computer Vision and Pattern Recognition

Citations in the corpus

Image-Based Recommendations on Styles and Substitutes (2015, background): “The system shares these characteristics with more common forms of recommender system, but does so on the basis of the appearance of the object, rather than metadata, reviews, or similar.”
Imbalanced Continual Learning with Partitioning Reservoir Sampling (2020, background): “There have been many works handling the vital problem of multi-label classification [59,32].”
Deep Sets (2017, methodology): “Speciﬁcally, we have Least Sq., a ridge regression model, MBRM (Feng , 2004), JEC (Makadia et al., 2008) and FastTag (Chen et al., 2013).”
Deep Convolutional Ranking for Multilabel Image Annotation (2013, methodology): “The work by Makadia et al. [25], which proposed a simple nearest-neighbor-based tag transfer approach, achieved signiﬁcant improvement over previous model-based methods.”
End-to-End Learning of Deep Visual Representations for Image Retrieval (2016, methodology): “This has proved useful for annotations as diverse as image-level tags (Makadia et al, 2008), GPS coordinates (Hays and Efros, 2008), or prominent object location (RodriguezSerrano et al, 2015).”
Multilabel Image Classification With Regional Latent Semantic Dependencies (2016, background): “With the available of the large-scale dataset and the enrichment of the data annotations, multi-label image classification has drawn lots of attentions [20, 21, 35].”
TagProp: Discriminative metric learning in nearest neighbor models for image auto-annotation (2009, unknown): “First, using the tag transfer method proposed in [17] with our own features we obtain results very similar to the original work.”
Nonparametric Method for Data-driven Image Captioning (2014, background): “Data-driven matching methods have shown to be very effective for a variety of challenging problems (Hays and Efros, 2008; Makadia et al., 2008; Tighe and Lazebnik, 2010).”
Learning the Relative Importance of Objects from Tagged Images for Retrieval and Cross-Modal Search (2012, background): “an appropriate word or distribution of words (Qi et al. 2009; Makadia et al. 2008; Quattoni et al. 2007; Loeff and Farhadi 2008; Bekkerman and Jeon 2007; Hardoon and ShaweTaylor 2003; Yakhnenko and Honavar 2009), the semantic space we discover also preserves the relative significance of the objects ”
Going Beyond Text: A Hybrid Image-Text Approach for Measuring Word Relatedness (2011, methodology): “For a global representation, features are often described using a continuous feature space, such as a color histogram in three different color spaces (RGB, HSV and LAB), or textures using Gabor and Haar wavelets (Makadia et al., 2008).”
A Thousand Frames in Just a Few Words: Lingual Description of Videos through Latent Topics and Sparse Object Stitching (2013, unknown): “[18] and TagProp [12], rely on large annotated sets to generate descriptions from similar samples.”
Deep Classifiers from Image Tags in the Wild (2015, methodology): “Our method Method Recall Precision N+ JEC [15] 29 28 250 TagProp σML [8] 35 46 266 SKL-CRM [17] 32 51 274 RLR (ours) 41 46 277”
Online Nonnegative Matrix Factorization With Robust Stochastic Approximation (2012, methodology): “To annotate a given image, the joint equal contribution model (JEC) [43] transfers keywords from the training images as an annotation in a greedy manner.”
CNN-RNN: A Unified Framework for Multi-label Image Classification (2016, background): “As a result, multi-label classification task is receiving increasing attention [12, 9, 24, 36].”
A Multi-View Embedding Space for Modeling Internet Images, Tags, and Their Semantics (2012, methodology): “As described in Section 5, we use the data-driven annotation scheme of Makadia et al. (2008), where tags are transferred from top fifty neighbors to the query in the latent space.”
Sparse Bayesian Multi-Task Learning (2011, background): “A problem that attracted considerable interest in recent years is to label an image with (text) keywords based on the features extracted from that image [26].”
Large scale image annotation: learning to rank with joint word-image embeddings (2010, methodology): “Retrieving the ﬁrst neighbor, at least on this dataset, is very important for achieving good accuracy, a phenomena that has been observed on other datasets, see e.g. Table 3 of Makadia et al. (2008).”
Image Classification Using Super-Vector Coding of Local Image Descriptors (2010, methodology): “The previous work includes SVMs using pyramid matching kernels [4], biologically-inspired models [5][6], and KNN methods [7][8][9].”
Im2Text and Text2Im: Associating Images and Texts for Cross-Modal Retrieval (2014, methodology): “In case of labels, we use the same set of annotations as in [10, 19, 35].”