Example-based synthesis of 3D object arrangements

By Matthew Fisher, Daniel Ritchie, Manolis Savva, Thomas A. Funkhouser, P. Hanrahan. 2012

413 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

ShapeNet: An Information-Rich 3D Model Repository (2015, methodology): “Mirroring this pattern, recent work in computer graphics has also applied similar approaches to speciﬁc problems in the synthesis of new shape variations [10] and new arrangements of shapes [6].”
The AdobeIndoorNav Dataset: Towards Deep Reinforcement Learning based Real-world Indoor Robot Visual Navigation (2018, background): “Existing 3D scenes datasets supporting the training of robot navigation can generally be categorized into two categories: 1) synthetic scenes [6], [7], [8], [9], [10], [11] 3D reconstructed scenes [12], [13], [14].”
Joint 2D-3D-Semantic Data for Indoor Scene Understanding (2017, background): “Although several RGB-D [6, 7, 16, 15] and a few 3D datasets [12, 13, 14, 1] have been developed to date, the majority of them are limited in the scale, diversity, and/or the number of modalities they provide.”
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).”
Semantically-enriched 3D models for common-sense knowledge (2015, methodology): “Real objects are composed of materials with properties that influence the appearance, density and texture of parts of the object and consequently their functionality (e.g., many chair seats are made from fabrics that are soft and comfortable to sit on).”
The Replica Dataset: A Digital Replica of Indoor Spaces (2019, methodology): “The Stanford Scenes [12] dataset consists of 130 scenes with 1,723 object instances.”
SceneNet RGB-D: 5M Photorealistic Images of Synthetic Indoor Trajectories with Ground Truth (2016, neutral): “We host the dataset at http://robotvault. bitbucket.io/scenenet-rgbd.html.”
IM2CAD (2016, positive): “Researchers have explored a variety of techniques to automatically compute CAD scene models using non-photographic means, e.g., using example based approaches [11], utilizing text descriptions [7], and optimizing for furniture arrangements in a given space [47, 28].”
PiGraphs (2016, methodology): “The domain of scenes has also seen its share of data-driven methods in the line of work addressing scene synthesis [Yu et al. 2011; Fisher et al. 2012; Xu et al. 2014].”
Understanding RealWorld Indoor Scenes with Synthetic Data (2015, background): “The idea of using synthetic scenes has existed in the past, in particular, [12], who released a number of scenes targetted towards the application of scene retrieval, which could potentially be used in the problem we are interested in.”
SceneGraphNet: Neural Message Passing for 3D Indoor Scene Augmentation (2019, background): “Alternatively, Bayesian Networks have been proposed to synthesize arrangements of 3D objects by modeling object co-occurence and placement statistics [4].”
SceneFormer: Indoor Scene Generation with Transformers (2020, background): “Yet other works condition on full 3D scans [15], or focus on fine-grained and smaller objects such as a table with multiple objects on it [12].”
SceneGrok (2014, methodology): “We then present examples of action map predictions on scanned test scenes, and on synthetic 3D scenes from a database assembled by prior work [Fisher et al. 2012].”
Scenic: a language for scenario specification and scene generation (2018, methodology): “A popular approach is to provide example tests, as in mutational fuzz testing [43] and example-based scene synthesis [10].”
Creating consistent scene graphs using a probabilistic grammar (2014, methodology): “2014], and automatic scene synthesis [Fisher et al. 2012].”
SceneNet: An annotated model generator for indoor scene understanding (2016, methodology): “To provide object intra-class variability, we sample from a broader library of objects e.g. ModelNet [5] and groups of objects such as Stanford Database [18].”