NUS-WIDE: a real-world web image database from National University of Singapore

By Tat-seng Chua, Jinhui Tang, Richang Hong, Haojie Li, Zhiping Luo, Yantao Zheng. 2009

3,342 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition, Evaluation & Benchmarks

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 Evaluation & Benchmarks

Citations in the corpus

Elucidating image-to-set prediction: An analysis of models, losses and datasets (2019, methodology): “The most widely used datasets for this task are adapted versions of object detection datasets (Pascal VOC [11], MS COCO [25]) or hashtag prediction datasets (NUS-WIDE [8]).”
A Multi-View Embedding Space for Modeling Internet Images, Tags, and Their Semantics (2012, methodology): “…of publications have reported better results with simple data-driven schemes based on retrieving database images similar to a query and transferring the annotations from those images (Chua et al., 2009; Guillaumin et al., 2009; Makadia et al., 2008; Verma and Jawahar, 2012; Wang et al., 2008).”
A Communication Efficient Vertical Federated Learning Framework (2019, methodology): “The NUS-WIDE dataset [18] consists of low-level images features and text tag features extracted from Flickr images.”
Imbalanced Continual Learning with Partitioning Reservoir Sampling (2020, methodology): “We further curate a sequential dataset from NUS-WIDE [17], containing 6 mutually exclusive and increasingly difficult tasks.”
Deep Convolutional Ranking for Multilabel Image Annotation (2013, methodology): “In our approach for multilabel image annotation, we adopted the architecture proposed in [20] as our basic framework and mainly focused on training the network with loss functions tailored for multi-label prediction tasks.”
Tencent ML-Images: A Large-Scale Multi-Label Image Database for Visual Representation Learning (2019, methodology): “Some widely used databases include Corel 5k [16] (including 4,999 images with 260 object categories), ESP Game [17] (including 20,770 images with 268 categories), IAPRTC-12 [18] (including 19,627 images with 291 categories), NUSWIDE [19] (including 270K images with 81 categories), MS COCO [20] (incl”
Heterogeneous Network Embedding via Deep Architectures (2015, methodology): “The second one is a heterogeneous dataset, which is referred to as NUS-WIDE [5].”
Pairwise Relationship Guided Deep Hashing for Cross-Modal Retrieval (2017, background): “NUS-WIDE (Chua et al. 2009): It is a real-world web image dataset containing 269648 instances, each including an image and its associated textual tags.”
Unsupervised Deep Hashing with Similarity-Adaptive and Discrete Optimization (2018, positive): “NUS-WIDE The NUS-WIDE dataset [4] consists of 269,648 images collected from Flickr.”
Central Similarity Quantization for Efficient Image and Video Retrieval (2019, methodology): “Datasets We use three image benchmark datasets, including ImageNet [30], NUS WIDE [5] and MS COCO [20].”
Deep Supervised Hashing for Fast Image Retrieval (2016, neutral): “(2) NUS-WIDE [1].”
Multimodal Manifold Analysis by Simultaneous Diagonalization of Laplacians (2015, positive): “NUS is a subset of the NUS-WIDE dataset [45] containing images (represented by 64-dimensional color histograms) and their text annotations (represented by 1,000-dimensional bags of words).”
M5Product: A Multi-modal Pretraining Benchmark for E-commercial Product Downstream Tasks (2021, methodology): “MS COCO [8], Flickr30K [27], INRIA-Websearch [31] and NUS-WIDE [32] with standard annotations are often used in the multi-modal and cross-modal retrieval tasks.”
Exploring Context and Content Links in Social Media: A Latent Space Method (2012, background): “elements in each feature vector is one [10].”
SCH-GAN: Semi-Supervised Cross-Modal Hashing by Generative Adversarial Network (2018, methodology): “We take 1% data of NUSWIDE dataset as the query set, and the rest as the retrieval database.”