Learning Everything about Anything: Webly-Supervised Visual Concept Learning

By Santosh Kumar Divvala, Ali Farhadi, Carlos Guestrin. 2014

325 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

Learning Transferable Visual Models From Natural Language Supervision (2021, background): “Of this line of work, Learning Everything about Anything: Webly-Supervised Visual Concept Learning (Divvala et al., 2014) has a notably similar ambition and goal as CLIP.”
Revisiting Unreasonable Effectiveness of Data in Deep Learning Era (2017, background): “In order to overcome the bottleneck, there have been recent efforts on visual representation learning using web-supervision [2, 5, 6, 9, 21, 23, 24, 27] or unsupervised [10, 11, 31, 32, 34, 42, 43] paradigms.”
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].”
Learning Visual Features from Large Weakly Supervised Data (2015, background): “This study is not the ﬁrst to explore alternatives to training convolutional networks on manually annotated datasets [6, 10, 41, 59].”
Webly Supervised Learning of Convolutional Networks (2015, background): “So, what we want is an approach that can learn visual representation from Flickr-like images.”
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 [46, 8, 10, 32, 49, 48].”
RegionCLIP: Region-based Language-Image Pretraining (2021, background): “The most relevant work is learning from natural language, such as image tags [3, 8, 12, 25, 28] and text descriptions [11, 23, 43, 53, 61].”
From captions to visual concepts and back (2014, background): “Determining the salient content requires not only knowing the contents of an image, but also deducing which aspects of the scene may be interesting or novel through commonsense knowledge [48, 4, 6].”
Grounding of Textual Phrases in Images by Reconstruction (2015, background): “There are also works that propose to train detectors for a wide range of concepts using image-level annotated data from web image search [17] and [26].”
iCaRL: Incremental Classifier and Representation Learning (2016, background): “[6] introduced systems that autonomously retrieve images from web resources and identifies relations between them, but they does not incrementally learn object classifiers.”
The Unreasonable Effectiveness of Noisy Data for Fine-Grained Recognition (2015, methodology): “4 is inspired by methods that propose to learn from web data [13, 8, 9, 60, 40, 29, 16] or reason about label noise [34, 57, 52, 47, 38].”
RECALL: Replay-based Continual Learning in Semantic Segmentation (2021, background): “Webly-Supervised Learning is an emerging paradigm in which large amounts of web data is exploited for learning CNNs [8, 10, 32].”
End-to-End Incremental Learning (2018, neutral): “Another relevant approach is learning concepts over time, in the form of lifelong [33] or never-ending [5, 7, 20] learning.”
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].”
Exploring Visual Relationship for Image Captioning (2018, background): “, actions or interactions) between objects are explored in [6,32] where each possible combination of semantic relation is taken as a visual phrase class and the visual relationship detection is formulated as a classification task.”
Don't just listen, use your imagination: Leveraging visual common sense for non-visual tasks (2015, background): “By visual common sense we do not mean visual models of commonly occurring interactions between objects [11] or knowledge of visual relationships between objects, parts and attributes [9, 50].”
Large-Scale Weakly-Supervised Pre-Training for Video Action Recognition (2019, background): “Dataset sources: Many prior approaches use Internet as a natural source of content [6, 12, 13, 18, 24, 25, 34, 37, 44, 46, 51, 58, 67] and the associated search queries, hashtags, or user-comments as labels to construct datasets.”
Computational Baby Learning (2014, methodology): “For the first category, existing image-based approaches [21] [2] [6] [3] iteratively used image search and detection results to cover more variations.”
Domain Adaptation for Visual Applications: A Comprehensive Survey (2017, methodology): “268, 269, 270, 95] or multi-view embedding [93, 110, 99, 111, 112, 113]. Hence, HDA/HTL can strongly beneﬁt from such cross-modal data representations. In the same spirit, webly supervised approaches [271, 272, 273, 274, 158, 159, 275] are also related to DA and HDA as is these approaches rely on co”