The MIR flickr retrieval evaluation

By Mark J. Huiskes, Michael S. Lew. 2008

1,948 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

Webly Supervised Joint Embedding for Cross-Modal Image-Text Retrieval (2018, background): “On the other hand, streams of images with noisy tags are readily available in datasets, such as Flickr-1M [21], as well as in nearly infinite numbers on the web.”
Self-Supervised Learning of Visual Features through Embedding Images into Text Topic Spaces (2017, neutral): “This includes classical image classification [5, 10], annotation [9, 17], and captioning [24, 23].”
Generative Adversarial Text to Image Synthesis (2016, background): “In comparison, natural language offers a general and ﬂexible interface for describing objects in any space of visual categories.”
Conditional Generative Adversarial Nets (2014, methodology): “For our experiments we use MIR Flickr 25,000 dataset [11], and extract the image and tags features using the convolutional model and language model we described above.”
MM-REACT: Prompting ChatGPT for Multimodal Reasoning and Action (2023, background): “For example, when determining if an image contains “people”, we may choose the image tagging model [7,16,19] and check if the predicted tag list contains “people”.”
Deep Compositional Captioning: Describing Novel Object Categories without Paired Training Data (2015, methodology): “We consider three different sources for unpaired text data to train the language model: (1) MSCOCO consists of all captions from the MSCOCO train set (2) Text from Image Description Corpora (CaptionTxt) consists of text data from other paired image and video description datasets: Flickr1M [13], Flic”
Efficient On-the-fly Category Retrieval Using ConvNets and GPUs (2014, methodology): “Adopting such an evaluation protocol also has the advantage that we are able to use the 1M images from the MIRFLICKR-1M dataset despite the fact that full annotations are not provided.”
Distance based similarity models for content based multimedia retrieval (2013, background): “, 1996], MIR Flickr [Huiskes and Lew, 2008], and 101 Objects [Fei-Fei et al.”
CIFDM: Continual and Interactive Feature Distillation for Multi-Label Stream Learning (2021, methodology): “The dataset MirFlickr has 1000 features and 38 labels.”
Improved Multimodal Deep Learning with Variation of Information (2014, methodology): “In this section, we evaluate our methods on MIR-Flickr database [11], which is composed of 1 million examples of image and their user tags collected from the social photo-sharing website Flickr.”
Multimodal semi-supervised learning for image classification (2010, methodology): “For the MIR Flickr data set we kept the tags that appear at least 50 times ( i.e . among at least 0.2% of the images), resulting in a vocabulary of 457 tags.”
Learning Discriminative Binary Codes for Large-scale Cross-modal Retrieval (2017, methodology): “5) Convergence Study: As DCH is solved by an iterative procedure in Algorithm 1, in this part we assess its convergence property by using 32 bits codes on Wiki and MIRFlickr datasets.”
Learning Robust Representations via Multi-View Information Bottleneck (2020, methodology): “The MIR-Flickr dataset (Huiskes & Lew, 2008) consists of 1M images annotated with 800K distinct user tags.”
Image Labeling on a Network: Using Social-Network Metadata for Image Classification (2012, methodology): “Focusing on the ﬁrst question we build on the intuition that images sharing similar tags and appearance are likely to have similar labels [9].”
Semantics-preserving hashing for cross-view retrieval (2015, neutral): “Wiki [12], MIRFlickr [7] and NUS-WIDE [2].”
A Decade Survey of Content Based Image Retrieval Using Deep Learning (2020, background): “National University of Singapore [46]; the Sun397 is a scene understanding dataset from 397 categories with more than one lakh images [49], [91]; and the MIRFlicker-1M [90] dataset consists of a million images downloaded from the social”
Block-Row Sparse Multiview Multilabel Learning for Image Classification (2016, background): “In order to evaluate the performance of the proposed block-row sparse MVML learning (F2L21F5 for short in this paper), we compared it with two baseline algorithms and three state-of-the-art algorithms on public datasets (e.g., MIRFLICKR [64] and NUS-WIDE [65]) in the MVML learning, in terms of the e”
Deep Compositional Cross-modal Learning to Rank via Local-Global Alignment (2015, methodology): “We evaluate the cross-modal ranking performance on two benchmark datasets, namely Pascal VOC’07 (Pascal07) [11] and Flickr8K [10].”