Characterizing structural relationships in scenes using graph kernels

By Matthew Fisher, Manolis Savva, P. Hanrahan. 2011

251 citations recorded by the source corpus.

Subfields: Computational Mechanics, Computer Graphics and Computer-Aided Design, 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 Computational Mechanics

In Computer Graphics and Computer-Aided Design

In Computer Vision and Pattern Recognition

Citations in the corpus

Continuous Scene Representations for Embodied AI (2022, background): “However, others propose to create 3D scene graphs that encode 3D relationships between the components of a scene [2, 12, 20, 51, 66].”
Scene Graph Generation by Iterative Message Passing (2017, methodology): “Scene graphs have been used in many downstream tasks such as image retrieval [18], 3D scene synthesis [4] and understanding [10], visual question answering [37], and automatic caption evaluation [1].”
Discovering objects and their relations from entangled scene representations (2017, methodology): “In graphics, a number of approaches have been used to capture contextual scene structure, such as energy models (Yu et al., 2011), graphical and mixture models (Fisher et al., 2011; 2012), stochastic grammars (Liu et al., 2014), and probabilistic programs (Talton et al., 2012; Yeh et al., 2012).”
SceneGrok (2014, methodology): “This work introduced the concept of human-centric interactions to the shape analysis literature which targets higher-level understanding of both shape collections [Mitra et al. 2013a; Mitra et al. 2013b] and 3D scenes [Fisher et al. 2011].”
Hallucinating Humans for Learning Robotic Placement of Objects (2012, unknown): “[30, 31] designed a context-based search engine using geometric cues and spatial relationships to find the proper object for a given scene.”
Learning part-based templates from large collections of 3D shapes (2013, methodology): “Finally, prior work [Fisher et al. 2011; Kalogerakis et al. 2012] on probabilistic modeling of shape variations requires manually annotated data sets with parts and correspondences already labeled.”
ConceptGraphs: Open-Vocabulary 3D Scene Graphs for Perception and Planning (2023, background): “3D scene graphs (3DSGs) address the second limitation by compactly and efficiently describing scenes with graph structures, with nodes representing objects and edges encoding inter-object relationships [44]– [48].”
Image retrieval using scene graphs (2015, methodology): “Our model draws inspiration from recent work in computer graphics that uses graph-based formulations to compare [20] and generate [7] scenes.”
Learning Object Arrangements in 3D Scenes using Human Context (2012, background): “There are also some previous works that consider human pose and activity recognition, and we describe them below.”
Configurable 3D Scene Synthesis and 2D Image Rendering with Per-pixel Ground Truth Using Stochastic Grammars (2017, background): “…2015; Bansal et al. 2016), semantic segmentation (Long et al. 2015; Noh et al. 2015; Chen et al. 2016), reasoning about object-supporting relations (Fisher et al. 2011; Silberman et al. 2012; Zheng et al. 2015; Liang et al. 2016), material recognition (Bell et al. 2013, 2014, 2015; Wu et al.…”
SceneGraphNet: Neural Message Passing for 3D Indoor Scene Augmentation (2019, background): “Early work in 3D scene modeling employed hand-engineered kernels and graph walks to retrieve objects from a database of 3D models that are compatible with the surrounding context of an input scene [3, 5].”
3DNN: 3D Nearest Neighbor (2014, unknown): “The graphics community has begun harvesting data from online repositories such as 3D Warehouse in an effort to better understand and model how objects are typically arranged in homes (Fisher and Hanrahan 2010; Fisher et al. 2011).”
Data-driven contextual modeling for 3D scene understanding (2016, background): “However, a key prerequisite for contextual scene analysis is that all objects are segmented and la-49 beled with semantic tags [9], which is apparently infea-50 sible for an unsegmented scene.”
Creating consistent scene graphs using a probabilistic grammar (2014, methodology): “2013], context-based object retrieval [Fisher and Hanrahan 2010], scene retrieval [Fisher et al. 2011], scene organization [Xu et al.”