Contextual Priming for Object Detection

By Antonio Torralba. 2003

953 citations recorded by the source corpus.

Subfields: Cognitive Neuroscience, Computer Vision and Pattern Recognition, Social Psychology

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Cognitive Neuroscience

In Computer Vision and Pattern Recognition

In Social Psychology

Citations in the corpus

Inside-Outside Net: Detecting Objects in Context with Skip Pooling and Recurrent Neural Networks (2015, neutral): “Global average pooling provides information about the entire image, similar to the features used for scene or image classification [25, 20].”
A MultiPath Network for Object Detection (2016, background): “However, as discussed, context is known to play an important role in object recognition [30].”
ParseNet: Looking Wider to See Better (2015, background): “In classical computer vision, before the current deep-learning revolution, such context was studied in detail [29, 23].”
LabelMe: A Database and Web-Based Tool for Image Annotation (2008, background): “Also, we show how to extend the dataset to automatically enhance object labels with WordNet, discover bject parts, recover a depth ordering of objects in a scene, and increase the number of labels using minimal user supervision and images from the web.”
Gather-Excite: Exploiting Feature Context in Convolutional Neural Networks (2018, background): “In computer vision, an idea that has often improved visual representations is to augment functions that perform local decisions with functions that operate on a larger context, providing a cue for resolving local ambiguities [39].”
Scene Graph Generation by Iterative Message Passing (2017, background): “Such representation not only offers contextual cues for fundamental recognition tasks [27, 29, 38, 39] but also provide values in a larger variety of high-level visual tasks [18, 44, 40].”
Geometric context from a single image (2005, background): “[29, 30] has been very Figure 1: Geometric context from a single image: ground (green), sky (blue), vertical regions (red) subdivided into planar orientations (arrows) and non-planar solid (’x’) and porous (’o’).”
Sharing Visual Features for Multiclass and Multiview Object Detection (2007, background): “Such global features can be used to do contextual priming of objects [28].”
An empirical study of context in object detection (2009, background): “Several recent computer vision approaches have demonstrated that the use of context improves recognition performance [4, 11, 14, 17, 24, 26, 32, 35, 39, 41].”
Modeling Visual Context is Key to Augmenting Object Detection Datasets (2018, background): “Relatively early, visual context has been modeled by computing statistical correlation between low-level features of the global scene and descriptors representing an object [17,18].”
REPAIR: Removing Representation Bias by Dataset Resampling (2019, background): “For example, contextual bias allows recognition algorithms to recognize objects by simply observing their environment [31].”
Objects in Context (2007, background): “In [9] [14] [25] [28], the relationship between context and object properties is based on the correlation between the statistics of low-level features across the image that contains the object, or even the whole object category.”
Semantic Jitter: Dense Supervision for Visual Comparisons via Synthetic Images (2016, methodology): “For RankSVM, we use Gist [40] and 30-bin Lab color histograms as the image features φ, following [28, 47]3, and validate K per method on held-out data.”
Object tracking: A survey (2006, background): “Recent work in the area of object recognition [Torralba 2003; Kumar and Hebert 2003] has shown that exploiting contextual information is helpful in recognition.”
Discovering object categories in image collections (2005, result): “We compare the two statistical models with a control global texture model, similar to those proposed for preattentive vision [22] and image retrieval [19].”
Modeling Scene and Object Contexts for Human Action Retrieval With Few Examples (2011, background): “Context has also been explored in several recent works for human action recognition [5], [10], [18]–[22].”
A Tree-Based Context Model for Object Recognition (2012, background): “Torralba [28] demonstrates that a global image feature called " gist " can predict the presence or absence of objects and their locations without running an object detector.”
Distilling Localization for Self-Supervised Representation Learning (2020, background): “Context plays an important role in recognizing objects [39].”
Building the gist of a scene: the role of global image features in recognition. (2006, background): “…real-world scenes (global contextual priming, Torralba 2003a) and, on the other hand, provide a formal instance of a feed-forward mechanism for scene context evaluation, for the guidance of attention and eye movements in the scene (Oliva et al., 2003; Torralba, 2003a,b; Torralba et al., 2006).”
Computational visual attention systems and their cognitive foundations: A survey (2010, methodology): “The contextual information is obtained from past search experiences in similar environments [Oliva et al. 2003; Torralba 2003b].”