Beyond Categories: The Visual Memex Model for Reasoning About Object Relationships

By Tomasz Malisiewicz, Alexei A. Efros. 2009

148 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

Context Encoders: Feature Learning by Inpainting (2016, methodology): “Visual Memex [29] used context to non-parametrically model object relations and to predict masked objects in scenes, while [6] used context to establish correspondences for unsupervised object discovery.”
The More You Know: Using Knowledge Graphs for Image Classification (2016, background): “[20] builds a graph of exemplars for different categories and uses the spatial relationships to perform contextual reasoning.”
Improving Generalization via Scalable Neighborhood Component Analysis (2018, background): “On the other hand, data association approaches [2,50,23,24] regard categories as data-driven entities emergent from connections between individual instances.”
Discovering states and transformations in image collections (2015, background): “Photo collections have also been mined for storylines [15] as well as spatial and temporal trends [18], and systems have been proposed for more general knowledge discovery from big visual data [21], [1], [3].”
A Tree-Based Context Model for Object Recognition (2012, unknown): “relationships based on different viewpoints [19].”
Unsupervised Visual Representation Learning by Graph-Based Consistent Constraints (2016, methodology): “Unlike other methods that use the nearest neighbor graphs to learn similarity functions [13,14], we use the graph to mine constraints for learning rich visual representations.”
FlowWeb: Joint image set alignment by weaving consistent, pixel-wise correspondences (2015, background): “Recent work exploits the connectivity within an image collection by deﬁning a graph over images (e.g. [15, 27, 47, 13]) or objects (e.g. [28, 20, 6]).”
The Visual Extent of an Object (2011, background): “…and Triggs 2005; Fulkerson et al. 2009; Gould et al. 2009; Shotton et al. 2009), 2D scene gist context (Oliva and Torralba 2001), 3D geometric context (Hoiem et al. 2008; Nedovi´c and Smeulders 2010), and semantic context (Malisiewicz and Efros 2009; Rabinovich et al. 2007; Singhal et al. 2003).”
Deep Reasoning with Knowledge Graph for Social Relationship Understanding (2018, methodology): “Representing extr/prior knowledge in the form of graph structure [ Schlichtkrull et al. , 2017; Lin et al. , 2017 ] and incorporating this structure for visual reasoning [ Malisiewicz and Efros, 2009; Zhu et al. , 2014; Teney et al. , 2017 ] has received increasingly attention.”
Visual Semantic Navigation using Scene Priors (2018, background): “…has been studied extensively in the computer vision literature (Torralba et al., 2003; Hoiem et al., 2005; Rabinovich et al., 2005; Divvala et al., 2009; Desai et al., 2009; Marszalek et al., 2009; Malisiewicz & Efros, 2009; Mottaghi et al., 2014; Zhu et al., 2015; Shrivastava & Gupta, 2016).”
3D Scene Graph: A Structure for Unified Semantics, 3D Space, and Camera (2019, background): “Another interesting example is that of Visual Memex [36] that leverages a graph structure to encode contextual and visual similarities between objects without the notion of categories, with the goal to predict the object class laying under a masked region.”
Graph-Based Discriminative Learning for Location Recognition (2013, background): “[17] [25] [18] This question of representation has analogues in more general object recognition problems, where many approaches regard objects as belonging to pre-defined categories (cars, planes, bottles, etc.), but other work represents objects more implicitly as structural relations between image”
Deep Learning for Generic Object Detection: A Survey (2018, background): “o other objects in the scene; Deep Learning for Generic Object Detection: A Survey 23 3.Scale context: Objects have a limited set of sizes relative to other objects in the scene. A great deal of work [34,56,76,181,189,216,203] preceded the prevalence of deep learning, however much of this work has n”
Context-based search for 3D models (2010, methodology): “Comparing models based on their tags has the same problems as using object categories in computer vision [Malisiewicz and Efros 2009].”
Data-driven visual similarity for cross-domain image matching (2011, methodology): “In particular, one can construct a visual memex graph (using the terminology from [Malisiewicz and Efros 2009]), whose nodes are images/subimages, and edges are various types of associations, such as visual similarity, context, etc.”
Representation Learning for Visual-Relational Knowledge Graphs (2017, background): “Recent years have witnessed a surge in reasoning about human-object, object-object, and object-attribute relationships [8, 12, 13, 17, 19, 28, 54, 56].”
The Inverted Multi-Index (2012, background): “More generally, they can be used within any computer vision task that involves fast near(est) neighbor retrieval or kernel density estimation (i.e. image classification [3], [4], understanding [5], image editing [6], etc.).”
COSTA: Co-Occurrence Statistics for Zero-Shot Classification (2014, methodology): “For example a chair is probably better described by contextual cues that are easier to recognize, such as indoors, table or desk, rather than by its composing parts that we may not even be able to define [7, 14].”