Author ' s personal copy The role of context in object recognition

By Aude Oliva, Antonio Torralba. 2007

How researchers see this paper

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

Citations in the corpus

Global-and-local attention networks for visual recognition (2018, neutral): “Decades of work on human attention have shown that our visual system uses at least two pathways to quickly guide attention to regions of interest [9]: A “global pathway” rapidly extracts a statistical summary of the whole scene in as little as a glance [10] while a complementary “local pathway” leve”
Contextual Action Recognition with R*CNN (2015, neutral): “More recently, Oliva and Torralba [22] study the contextual associations of objects with their scene and link various forms of context cues with computer vision.”
Scene Graph Generation by Iterative Message Passing (2017, positive): “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].”
An empirical study of context in object detection (2009, neutral): “Many different sources of context have been discussed in the literature [2, 29, 38] and others are proposed here (see Table 1 for summary).”
Modeling Visual Context is Key to Augmenting Object Detection Datasets (2018, positive): “Obviously, modeling correctly object appearances is important, but it is also well-known that visual context provides important cues for recognition, both for computer vision systems and for humans [1].”
Graph R-CNN for Scene Graph Generation (2018, neutral): “The idea of using context to improve scene understanding has a long history in computer vision [16,27,28,30].”
Auto-Context and Its Application to High-Level Vision Tasks and 3D Brain Image Segmentation (2010, unknown): “C ONTEXT and high-level information plays a vital role in object recognition and scene understanding [2], [28], [45].”
Deep Hierarchies in the Primate Visual Cortex: What Can We Learn for Computer Vision? (2013, unknown): “Interestingly, objects also help to recognize the context and context may be defined on a crude statistical level [124].”
Full DOF tracking of a hand interacting with an object by modeling occlusions and physical constraints (2011, unknown): “Given the significant role of context in human visual recognition [18], several researchers have attempted to exploit contextual constraints in solving computer vision problems.”
THINGS: A database of 1,854 object concepts and more than 26,000 naturalistic object images (2019, unknown): “…knowledge to The need for large-scale systematic sampling of object concepts and naturalistic object images In recent years, in the fields of psychology, neuroscience and computer science there has been a growing interest in utilizing a wide range of object concepts and naturalistic images [1–5].”
Learned Contextual Feature Reweighting for Image Geo-Localization (2017, unknown): “Contextual information has been widely used for object recognition [39, 42, 64, 8, 13, 27].”
Single-Pedestrian Detection Aided by Multi-pedestrian Detection (2013, unknown): “A review of context in object recognition is provided in [55].”
Learning Spatial Context: Using Stuff to Find Things (2008, unknown): “The role of context in object recognition has become an important topic, due both to the psychological basis of context in the human visual system [10] and to the striking algorithmic improvements that “visual context” has provided [11].”
Spatial pyramid matching (2009, unknown): “The second reason is an improved understanding of context (Hoiem et al., 2005; Oliva and Torralba, 2007), or the way that global image appearance and geometry inﬂuence the perception of individual objects in the scene.”
BlobGAN: Spatially Disentangled Scene Representations (2022, unknown): “Indeed, it was long clear, both in human vision [9, 26] and in computer vision [91, 54, 56, 27, 19], that understanding and modeling objects within the context of a scene is of the utmost importance.”
Teaching Compositionality to CNNs (2017, unknown): “It is well known that context plays a major role in visual recognition, both in human and artificial vision systems [9, 26, 5].”
Finding Tiny Faces (2016, unknown): “Though this observation is quite intuitive and highly explored in computer vision [16, 21], it has been notoriously hard to quantifiably demonstrate the benefit of context in recognition [4, 6, 22].”