Context based object categorization: A critical survey

By Wenming Tang, Yuanhao Gong, Guoping Qiu. 2010

592 citations recorded by the source corpus.

The goal of object categorization is to locate and identify instances of an object category within an image. Recognizing an object in an image is difficult when images include occlusion, poor quality, noise or background clutter, and this task becomes even more challenging when many objects are present in the same scene. Several models for object categorization use appearance and context information from objects to improve recognition accuracy. Appearance information, based on visual cues, can successfully identify object classes up to a certain extent. Context information, based on the interaction among objects in the scene or global scene statistics, can help successfully disambiguate appearance inputs in recognition tasks. In this work we address the problem of incorporating different types of contextual information for robust object categorization in computer vision. We review different ways of using contextual information in the field of object categorization, considering the most common levels of extraction of context and the different levels of contextual interactions. We also examine common machine learning models that integrate context information into object recognition frameworks and discuss scalability, optimizations and possible future approaches.

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

Relation Networks for Object Detection (2017, neutral): “We refer readers to [16] for a more detailed survey.”
Multi-Scale Context Aggregation by Dilated Convolutions (2015, neutral): “Semantic segmentation is challenging because it requires combining pixel-level accuracy with multi-scale contextual reasoning (He et al., 2004; Galleguillos & Belongie, 2010).”
Dilated Residual Networks (2017, positive): “Since contextual information is important in disambiguating local cues [3], such reduction in receptive field is an unacceptable price to pay for higher resolution.”
Neural Motifs: Scene Graph Parsing with Global Context (2017, methodology): “Our approach is most closely related to work that models object co-occurrence using graphical models to combine many sources of contextual information [30, 11, 23, 10].”
An empirical study of context in object detection (2009, neutral): “It also may indicate the presence and location (spatial context) of other objects and materials [10, 11, 12, 37].”
Visual Relationship Detection with Internal and External Linguistic Knowledge Distillation (2017, unknown): “object cooccurrence, spatial relationships) to improve visual tasks [30, 27, 21, 14, 4, 5, 15].”
A survey on face detection in the wild: Past, present and future (2015, unknown): “There has been some recent work on context based object categorization [214] and visual tracking [215].”
Contextual Object Detection Using Set-Based Classification (2012, unknown): “Contextual relationships can be represented in terms of the relations between local image patches [8,9], objects [10,2,3], coarse scene characteristics [11], local background regions [6] and the geometric layout of the scene [12] (see [13] for a review).”
Recent Advances in Deep Learning for Object Detection (2019, background): “For traditional object detection algorithms, there have been several efforts exploring context [126], but for object detection based on deep learning, context has not been extensively explored.”
Harmony Potentials (2011, background): “Several authors have noted the importance of context to obtain good classification (Oliva and Torralba 2007; Galleguillos and Belongie 2010).”
A Survey of Recent Advances in Face Detection (2010, unknown): “There has been some recent work on context based object categorization [22] and visual tracking [111].”
Contextualizing Object Detection and Classification (2015, unknown): “In the early work of Galleguillos and Belongie [20], the context refers to three main types of contextual information that can be exploited in computer vision: (1) the semantic context which refers to the likelihood of an object being found in some scenes but not in others, and from the point of vie”
Teaching Compositionality to CNNs (2017, background): “It is well known that context plays a major role in visual recognition, both in human and artificial vision systems [9, 26, 5].”