NEIL: Extracting Visual Knowledge from Web Data

By Xinlei Chen, Abhinav Shrivastava, Abhinav Kumar Gupta. 2013

497 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

Learning Video Representations from Textual Web Supervision (2020, methodology): “The most commonly-used approach is to use image search results, and to label each image with the query that was used to ﬁnd it [65, 15, 56, 6, 11, 13, 21, 10, 18, 35].”
Webly Supervised Learning of Convolutional Networks (2015, methodology): “We obtained three lists of categories from ImageNet Challenge [7], SUN database [8] and NEIL knowledge base [2].”
OK-VQA: A Visual Question Answering Benchmark Requiring External Knowledge (2019, background): “Several knowledge bases have been created using visual data or for visual reasoning tasks [53, 9, 11, 39, 56, 55].”
VQA: Visual Question Answering (2015, background): “However, it includes as building blocks several components that the CV, NLP, and KR [4, 6, 25, 29, 3] communities have made significant progress on during the past few decades.”
iCaRL: Incremental Classifier and Representation Learning (2016, background): “…standard classification loss, which encourages improvements of the feature representation that allow classifying the newly observed classes well, it also contains the distillation loss, which ensures that the discriminative information learned previously is not lost during the new learning step.”
The Unreasonable Effectiveness of Noisy Data for Fine-Grained Recognition (2015, methodology): “While we will provide all urls to the research community, note that this data acquisition is quite feasible for most researchers, and has been done in the past for similarly-sized [9] and even larger collection efforts [62, 65].”
The More You Know: Using Knowledge Graphs for Image Classification (2016, background): “Learning knowledge graphs [38, 4, 29] and using graphs for visual reasoning [39, 20] has recently been of interest to the vision community.”
End-to-End Incremental Learning (2018, neutral): “, [5], or rely on a fixed representation, e.”
Incremental Learning of Object Detectors without Catastrophic Forgetting (2017, background): “They are: (i) restricted to learning with a fixed data representation [9], or (ii) keep all the collected data to retrain the model [8].”
Training Deep Neural Networks on Noisy Labels with Bootstrapping (2014, background): “, 2010) and never ending image learning (NEIL) (Chen et al., 2013; 2014) are lifelong-learning systems for language and image understanding, respectively.”
A Wholistic View of Continual Learning with Deep Neural Networks: Forgotten Lessons and the Bridge to Active and Open World Learning (2020, background): “Here, the latter is adopted in the spirit of the older “never-ending learning” systems for language and images (NELL and NEIL)(Carlson et al., 2010; Chen et al., 2013; Mitchell et al., 2015), which take a more “traditional” AI approach.”
Explicit Knowledge-based Reasoning for Visual Question Answering (2015, methodology): “ation/crowd-sourcing (e.g., DBpedia [2], Freebase [6] and Wikidata [39]), or by automatically extracting from unstructured/semistructured data (e.g., YAGO [22, 28], OpenIE [3, 14, 15], NELL [9], NEIL [10, 11]). The KB we use here is DBpedia, which contains structured information extracted from Wikip”