Understanding Indoor Scenes Using 3D Geometric Phrases

By Wongun Choi, Yu-Wei Chao, Caroline Pantofaru, Silvio Savarese. 2013

187 citations recorded by the source corpus.

Subfields: Aerospace Engineering, Computer Vision and Pattern Recognition, Geology

How researchers see this paper

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

In Aerospace Engineering

In Computer Vision and Pattern Recognition

In Geology

Citations in the corpus

Visual Genome: Connecting Language and Vision Using Crowdsourced Dense Image Annotations (2016, background): “, 2014), and 3D spatial geometry between objects has helped object detection (Choi et al., 2013).”
Monocular Object Instance Segmentation and Depth Ordering with CNNs (2015, background): “This yields instance segmentation as well as depth ordering, however, segmentation is typically limited to a simple cuboid-like object shape.”
"What Happens If..." Learning to Predict the Effect of Forces in Images (2016, background): “There is a considerable body of work on scene understanding in the vision literature, e.g., [26,9,41,21,13,32,4,22,43].”
Visual Relationship Detection with Language Priors (2016, methodology): “nce models for visual phrases has shown to improve individual object detection, i.e. detecting \a person riding a horse" improves the detection and localization of \person" and \horse" [6,41]. Unlike our model, all previous work has attempted to detect only a handful of visual relatio”
Semantically-enriched 3D models for common-sense knowledge (2015, background): “The impact of gravity can be felt continuously by people and influences the structure of objects in the world.”
Joint 3D Object and Layout Inference from a Single RGB-D Image (2015, background): “Towards this goal, a number of holistic models have been proposed which take into account the relationship between objects (often represented as cuboids) and/or layout elements in the scene [4,22,37,45].”
Knowledge-Embedded Routing Network for Scene Graph Generation (2019, background): “Over the past decade, a series of works were dedicated to recognizing spatial relationships [9, 11, 5] like “above”, “below”, “inside”, and “around”, and to exploring using these relationships to improve various vision tasks such as object recognition [9], detection [8], and segmentation [11].”
Holistic++ Scene Understanding: Single-View 3D Holistic Scene Parsing and Human Pose Estimation With Human-Object Interaction and Physical Commonsense (2019, background): “(iii) Joint estimation of the room layout and 3D objects with contexts [40, 54, 7, 52, 62].”
CubeSLAM: Monocular 3-D Object SLAM (2018, methodology): “Therefore, we compared with two public approaches, SUN primitive [10] and 3D Geometric Phrases (3dgp) [46].”
Know More Say Less: Image Captioning Based on Scene Graphs (2019, background): “Earlier approaches explore speciﬁc types of relationships (e.g., spatial relations such as “below”, “above”) and often utilize them to facilitate other tasks [48], [49].”
3DNN: 3D Nearest Neighbor (2014, methodology): “As an additional baseline, we measure the performance of 3DGP Choi et al. (2013) for the task object detection and segmentation.”
3D Scene Graph: A Structure for Unified Semantics, 3D Space, and Camera (2019, background): “Grammars that combine geometry, affordance and appearance have been used toward holistic scene parsing in images, where information about the scene and objects is captured in a hierarchical tree structure [11, 48, 21, 19].”
Factorizable Net: An Efficient Subgraph-based Framework for Scene Graph Generation (2018, background): “In the early stage, most of the works targeted on using speciﬁc types of visual relations, such as spatial relations [5, 10, 14, 20, 26, 30] and actions ( i . e . interactions between objects) [1, 9, 11, 16, 19, 46, 47, 50, 56, 57, 60].”
FPM: Fine Pose Parts-Based Model with 3D CAD Models (2014, background): “vision community in addressing object detection [8,13] and 3D scene understanding [12,24,4,33,16], fine pose estimation is becoming more and more approachable.”
Semantic Image Interpretation - Integration of Numerical Data and Logical Knowledge for Cognitive Vision (2018, unknown): “Other methods for global SII graph construction are based on the minimization of an energy [19, 57, 17, 15] or a loss [31, 107] functions.”
DeepContext: Context-Encoding Neural Pathways for 3D Holistic Scene Understanding (2016, background): “e context is usually manually deﬁned as a unary term on a single object, pairwise term between a pair of objects to satisfy certain functionality [23,46], or a more complicated hierarchy architecture [6,24,49]. The learned context models are usually applied on a large set of object hypotheses genera”
School of Electrical and Computer Engineering Oklahoma State University , Stillwater , OK 74078 , USA (methodology): “The methods in the first group are RGB-based, which rely on extracting lines, edges, and orthogonal vanishing points from RGB images [9, 5].”
Are Cars Just 3D Boxes? Jointly Estimating the 3D Shape of Multiple Objects (2014, background): “Recently, more fine-grained, geometric representations have been successfully applied to the understanding of indoor scenes [7, 9, 38], where a box-like room layout is populated with furniture objects.”
Dark, Beyond Deep: A Paradigm Shift to Cognitive AI with Humanlike Common Sense (2020, background): “erating the best possible conﬁguration in terms of functionality, physics, and object relationships. This line of work has mostly been studied separately from recognition and semantics until recently [31,32,33,34,35,36,37,38]; see Fig.6[36] for an example. The idea of reconstruction as a “cognitive ”