Scene Generation & Artificial Intelligence

Notes:

Scene generation is a field within computer graphics and computer vision that focuses on generating synthetic images or videos that depict realistic scenes. Artificial intelligence (AI) can be used in scene generation in a number of ways, such as:

To generate 3D models of objects and scenes, using techniques such as deep learning and computer vision.
To control the lighting and shading in a scene, using techniques such as global illumination and physically-based rendering.
To animate objects and characters in a scene, using techniques such as inverse kinematics and motion capture.
To generate natural language descriptions of a scene, using natural language processing and generative modeling.

AI can be used to automate and improve many of the tasks involved in scene generation, such as object recognition, scene understanding, and image synthesis. This can enable the generation of more realistic and compelling scenes, and can also make the scene generation process more efficient and scalable.

3D reasoning refers to the ability of an artificial intelligence (AI) system to understand and reason about objects and scenes in a three-dimensional (3D) space. This can include tasks such as object recognition, scene understanding, and spatial reasoning.
Narrative visualization is the process of using visual elements, such as charts, graphs, and illustrations, to tell a story or convey information in a clear and engaging way. It is often used to present complex data or information in a way that is easy to understand and remember.
Scene generation is the process of creating a computer-generated image or animation that depicts a scene or environment. This can include tasks such as creating realistic 3D models of objects, characters, and environments, as well as simulating lighting, shading, and other visual effects.
Story visualization is the process of using visual elements, such as charts, graphs, and illustrations, to present a story or narrative. It is often used to present complex information or data in a way that is easy to understand and remember. It can also be used to create interactive visualizations that allow the user to explore and interact with the story in a more engaging way.
Scene generator is a software that generates computer-generated images or animations that depict a scene or environment. This can include tasks such as creating realistic 3D models of objects, characters, and environments, as well as simulating lighting, shading, and other visual effects. Scene generators can be used in fields such as animation, gaming, and virtual reality.
Text-to-image is a process of converting text-based descriptions of an image into a digital image. This can be done using machine learning algorithms that are trained on a dataset of images and their corresponding text descriptions. The resulting images can be used in fields such as computer vision, gaming, and virtual reality.
Text-to-scene is a process of converting a text description of a scene into a computer-generated image or animation that depicts that scene. This can include tasks such as creating realistic 3D models of objects, characters, and environments, as well as simulating lighting, shading, and other visual effects.
Text-To-Scene Conversion System (TTSCS) is a type of software that converts text descriptions of a scene into a computer-generated image or animation that depicts that scene. TTSCS is a type of Text-to-Scene technology, it can include a set of techniques, algorithms, and tools that can be used to generate realistic 3D environments, characters, and objects based on text descriptions. It can be used in fields such as computer vision, gaming, and virtual reality.
Text-to-scene system is a type of software that converts text-based descriptions of a scene into a computer-generated image or animation that depicts that scene. This can include tasks such as creating realistic 3D models of objects, characters, and environments, as well as simulating lighting, shading, and other visual effects. Text-to-scene systems can be used in fields such as computer vision, gaming, and virtual reality.
Text2scene is a specific type of text-to-scene system, it’s a name of a framework or model that can be used to convert text descriptions into 3D scenes. Text2scene can include a set of techniques, algorithms, and tools that can be used to generate realistic 3D environments, characters, and objects based on text descriptions. It can be used in fields such as computer vision, gaming, and virtual reality.

Resources:

uvavision/text2scene .. generating compositional scenes from textual descriptions

Know more say less: Image captioning based on scene graphs
X Li, S Jiang – IEEE Transactions on Multimedia, 2019 – ieeexplore.ieee.org
… IMAGE captioning is the task of automatically describing the content of images with natural language sentences, which has received increasing attention in the field of multimedia and artificial intelligence [1]–[7]. This task … We formulate the scene generation problem as follows …

VERIFAI: A toolkit for the design and analysis of artificial intelligence-based systems
T Dreossi, DJ Fremont, S Ghosh, E Kim… – arXiv preprint arXiv …, 2019 – arxiv.org
… In: Thirty-Second AAAI Conference on Artificial Intelligence (2018) 2. Alur, R., Henzinger, TA: Logics and models of real time: A survey … Fremont, DJ, Yue, X., Dreossi, T., Ghosh, S., Sangiovanni-Vincentelli, AL, Seshia, SA: Scenic: Language-based scene generation …

A layer-based sequential framework for scene generation with gans
MO Turkoglu, W Thong, L Spreeuwers… – arXiv preprint arXiv …, 2019 – arxiv.org
… Scene generation problem has been studied extensively in (Hertzmann et al. 2001; Tappan 2008; Chen et al … Copyright c 2019, Association for the Advancement of Artificial Intelligence (www.aaai.org). All rights reserved. Figure 1: The proposed image generation process …

VerifAI: A Toolkit for the Formal Design and Analysis of Artificial Intelligence-Based Systems
T Dreossi, DJ Fremont, S Ghosh, E Kim… – … on Computer Aided …, 2019 – Springer
… In: 27th International Joint Conference on Artificial Intelligence (IJCAI) (2018)Google Scholar. 10 … Fremont, DJ, Dreossi, T., Ghosh, S., Yue, X., Sangiovanni-Vincentelli, AL, Seshia, SA: Scenic: a language for scenario specification and scene generation …

Scenic: a language for scenario specification and scene generation
DJ Fremont, T Dreossi, S Ghosh, X Yue… – Proceedings of the 40th …, 2019 – dl.acm.org
Scenic: A Language for Scenario Specification and Scene Generation Daniel J. Fremont University of California, Berkeley USA dfremont@berkeley.edu … Scenic: A Language for Scenario Specification and Scene Generation PLDI ’19, June 22–26, 2019, Phoenix, AZ, USA …

Space Vector Generation for 3D Scenes from Text Descriptions
C Wang, B Tang, S Cao – … on Artificial Intelligence and Virtual Reality, 2019 – dl.acm.org
… Artificial intelligence ?Natural language processing ?Information extraction Keywords Test-to-spatial vector; natural language; 3D scene; neural network. 1. INTRODUCTION The use of speech and text descriptions to generate scenes in 3D scene generation saves time in …

Genesis: Generative scene inference and sampling with object-centric latent representations
M Engelcke, AR Kosiorek, OP Jones… – arXiv preprint arXiv …, 2019 – arxiv.org
… We increase the GECO step size by a constant factor when the reconstruction constraint is satisfied to accelerate training. 4.2 Scene Generation Unlike previous works, GENESIS has an autoregressive prior to capture intricate dependencies be- tween scene components …

Artificial Intelligence in Intelligent Tutoring Robots: A Systematic Review and Design Guidelines
J Yang, B Zhang – Applied Sciences, 2019 – mdpi.com
… Sci. 2019, 9(10), 2078; https://doi.org/10.3390/app9102078. Review. Artificial Intelligence in Intelligent Tutoring Robots: A Systematic Review and Design Guidelines. by Jinyu Yang 1 and Bo Zhang 2 … 3. Artificial Intelligence Techniques for Designing Intelligent Tutor Robots …

A Scene Division Method Based on Theme
F Yang, Z Yuan, X Cheng – 2019 – pdfs.semanticscholar.org
… It can realize the spatial placement of simple geometry, which is also a significant research result in the field of natural language processing and artificial intelligence … Finally, scene generation involves the production and presentation of 3D scenes or animations …

COSMO: contextualized scene modeling with boltzmann machines
I Bozcan, S Kalkan – Robotics and Autonomous Systems, 2019 – Elsevier
Skip to main content …

Greek language object representation scene system: a text interface for designing X3D objects and X3D scenes
GA Krikos, NN Karanikolas, G Miaoulis… – Proceedings of the 23rd …, 2019 – dl.acm.org
… CCS CONCEPTS Computing methodologies ? Computer graphics; Computing methodologies ? Artificial intelligence ? Natural language processing … Realtime Automatic 3D Scene Generation from Natural Language Voice and Text Descriptions …

A survey of 3D indoor scene synthesis
SH Zhang, SK Zhang, Y Liang, P Hall – Journal of Computer Science and …, 2019 – Springer
Page 1. Zhang SH, Zhang SK, Liang Y et al. A survey of 3D indoor scene synthesis. JOURNAL OF COMPUTER SCIENCE AND TECHNOLOGY 34(3): 594–608 May 2019. DOI 10.1007/s11390-019-1929-5 A Survey of 3D Indoor Scene Synthesis …

Generating a Ball Sport Scene in a Virtual Environment.
J Choi, S Kim, S Kim, S Kang – KSII Transactions on Internet …, 2019 – search.ebscohost.com
… [28] used the Mode-Adaptive Neural Networks to generate natural locomotion motions of quadrupedal animals. There have also been researches that use artificial intelligence to generate animations of interaction between characters and objects … 3. Sports Scene Generation …

Synthetic Video Generation for Robust Hand Gesture Recognition in Augmented Reality Applications
V Jain, S Aggarwal, S Mehta… – arXiv preprint arXiv …, 2019 – arxiv.org
… [7] Mehmet Ozgur Turkoglu, William Thong, Luuk Spreeuwers, and Berkay Kicanaoglu. A layer-based sequential framework for scene generation with gans. In Thirty-Third AAAI Confer- ence on Artificial Intelligence (AAAI-19), 2019 …

CriSGen: Constraint-based Generation of Critical Scenarios for Autonomous Vehicles
A Nonnengart, M Klusch, C Müller – pdfs.semanticscholar.org
… Andreas Nonnengart, Matthias Klusch, and Christian Müller German Research Center for Artificial Intelligence, Saarbrücken, Germany {firstname.lastname}@dfki.de Abstract … Fig. 1. Overview of the CriSGen approach for critical scene generation …

Teaching Computers to Teach Themselves: Synthesizing Training Data based on Human-Perceived Elements
J Little – 2019 – digitalcommons.bowdoin.edu
… Computer Science. First Advisor. Eric Chown. Abstract. Isolation-Based Scene Generation (IBSG) is a process for creating synthetic datasets made to train machine learning detectors and classifiers … Since. Included in. Artificial Intelligence and Robotics Commons. COinS …

Aesthetics, Artificial Intelligence, and Search-based Art
CG Johnson – 2019 – kar.kent.ac.uk
… Citation for published version Johnson, Colin G. (2019) Aesthetics, Artificial Intelligence, and Search-based Art. In: Romero, Juan and Machado, Penousal and Greenfield, Gary, eds … Page 2. Chapter 1 Aesthetics, Artificial Intelligence, and Search-based Art Colin G. Johnson …

CommonGen: A Constrained Text Generation Dataset Towards Generative Commonsense Reasoning
BY Lin, M Shen, Y Xing, P Zhou, X Ren – arXiv preprint arXiv:1911.03705, 2019 – arxiv.org
… Recall the illustrative example in Figure 1, even such a simple scene generation process needs pretty much common- sense knowledge like: 1) “apples grow in trees”; 2) “bags are containers that you can put some- thing in”; 3) “you usually pick something and then place it in a …

Pscenegan: Multi-Domain Particular Scenes Generation Based on Conditional Generative Adversarial Networks
LL Jia, XY Lv, YJ Cao, C Yang, XX Li, J Li – IEEE Access, 2019 – ieeexplore.ieee.org
… In this paper, we propose a multi-domain particular scene generation model named PSceneGAN (Particular Scene Generative Adversar- ial Nets) that is a novel dual-condition GAN … Therefore, we test our scene generation image using two approaches …

Vision and Language: from Visual Perception to Content Creation
T Mei, W Zhang, T Yao – arXiv preprint arXiv:1912.11872, 2019 – arxiv.org
… I. INTRODUCTION Computer Vision (CV) and Natural Language Processing (NLP) are two most fundamental disciplines under a broad area of Artificial Intelligence (AI) … Top: single object generation. Bottom: multiple-objects scene generation …

Habitat: A platform for embodied ai research
M Savva, A Kadian, O Maksymets… – Proceedings of the …, 2019 – openaccess.thecvf.com
… Abstract We present Habitat, a platform for research in embodied artificial intelligence (AI) … Scene graphs allow us to compose 3D environments through procedural scene generation, editing, or programmatic manipulation. Rendering engine …

Time Estimation as Critical Factor of Software Failure: A Systematic Literature Review Protocol
M Turakhia, U Nandu, P Shah… – i-Manager’s Journal …, 2019 – search.proquest.com
… In future, the authors plan to introduce complex Artificial Intelligence algorithms to this research work so as to make it learn the usual demands and imaginations for … The best part of “Speech to 3D Scene Generation” is that it is meant to be used by any person who can speak …

Illustration Design Teaching Mode Based on Virtual Wall Painting Technology
L Zhang – International Journal of Emerging …, 2019 – … journals.publicknowledgeproject.org
… education generated by sufficient utilization of 3D virtual scene is an important direction of VR technology in the educational field [1]. 3D virtual scene generation technology synthesizes computer graphic (CG), simulation technology, artificial intelligence and display technique …

Research on Spatial Conceptual Modeling of Natural Language Processing Based on Deep Learning Algorithms
J Wang – Journal of Physics: Conference Series, 2019 – iopscience.iop.org
… attention in academia is because of its breakthrough achievements in a series of important tasks of artificial intelligence [3]. At … modeling system mainly involves three aspects: the natural language understanding technology of the front end, the scene generation technology of …

Machine Imagination: A Step Toward the Construction of Artistic World Through Storytelling
STS Bukhari, A Kanwal, WM Qazi – … Trends and Advances in Wireless and …, 2019 – Springer
… With the continuous recall, AIA improved its scene generation, and results can be seen in Table 18.1. Table 18.1 Scene generation results. No … In Android epistemology (pp. 167–182). Cambridge, MA: MIT Artificial Intelligence Lab.Google Scholar. 24. Matari?, MJ (1990) …

Multimodal Intelligence: Representation Learning, Information Fusion, and Applications
C Zhang, Z Yang, X He, L Deng – arXiv preprint arXiv:1911.03977, 2019 – arxiv.org
… Abstract—Deep learning has revolutionized speech recognition, image recognition, and natural language processing since 2010, each involving a single modality in the input signal. However, many applications in artificial intelligence involve more than one modality …

Synthesizing Attributes with Unreal Engine for Fine-grained Activity Analysis
TS Kim, M Peven, W Qiu, A Yuille… – 2019 IEEE Winter …, 2019 – ieeexplore.ieee.org
… 2.1. Synthetic Data Generation with Unreal Engine Our synthetic data generation system is based on Unreal Engine (UE) including four modules. The first module is the scene generation module. It con- tains 3D assets and scene generation code to layout 3D ob- jects …

An Advanced Integrated Workflow for Automated and Quantitative Interpretation of Digital Outcrop Analogues
M Etchebes, A Bounaim, T Brenna… – First EAGE Workshop on …, 2019 – earthdoc.org
… technologies originally developed for automated seismic interpretation and further enriched with recent advances in 3D scene generation, visualization and … to new cloud-based platforms for data access and for generating training data to more artificial intelligence and machine …

A qualitative and localized evaluation for 3D indoor scene synthesis
H Liu – 2019 – summit.sfu.ca
… Chapter 1 Introduction Automatic scene generation is an ascendant line of work in computer graphics, particularly after the renaissance of Virtual reality and its descendants, Augmented reality (AR) and Mixed reality (MR) … Here, 3D scene generation plays an important role …

Story Envisioning Framework for Visualized Collective Storytelling in Conversation
Q Zhang, MS Mirzaei, HH Huang, T Nishida – International Conference on …, 2019 – Springer
… Chang, A., Monroe, W., Savva, M., Potts, C., Manning, CD: Text to 3D scene generation with rich lexical grounding (2015). arXiv preprint: arXiv:1505.06289. 15. Johnson, WL, Valente, A.: Tactical language and culture training systems: using artificial intelligence to teach Foreign …

Structured agents for physical construction
V Bapst, A Sanchez-Gonzalez, C Doersch… – arXiv preprint arXiv …, 2019 – arxiv.org
… 1. Introduction Humans are a “construction species”—we build forts out of couch cushions as children, pyramids in our deserts, and space stations that orbit hundreds of kilometers above our heads. What do artificial intelligence (AI) agents need to do these sorts of things …

Faster attend-infer-repeat with tractable probabilistic models
K Stelzner, R Peharz, K Kersting – … Conference on Machine …, 2019 – proceedings.mlr.press
… Deriving meaningful representations from data with inherent structure is a key problem in machine learning and artificial intelligence … model is Attend-Infer-Repeat (AIR) (Es- lami et al., 2016), which incorporates VAEs as object models within a scene generation process and …

Normalized Convolution Network and Dataset Generation for Refining Stereo Disparity Maps
D Cranston, F Skarfelt – 2019 – diva-portal.org
Page 1. Master of Science Thesis in Electrical Engineering Department of Electrical Engineering, Linköping University, 2019 Normalized Convolution Network and Dataset Generation for Refining Stereo Disparity Maps Filip Skarfelt and Daniel Cranston Page 2 …

Research on Key Technologies of Network Security Situational Awareness for Attack Tracking Prediction
G Kou, S Wang, G Tang – Chinese Journal of Electronics, 2019 – IET
… (1. Artificial Intelligence Research Center, National Innovation … Meanwhile, through the discussion of state transition scene which may appear in attack, it weakens the effect of report failure and repeated alert on scene generation, thus avoiding the appearance of return circuit …

Planit: Planning and instantiating indoor scenes with relation graph and spatial prior networks
K Wang, YA Lin, B Weissmann, M Savva… – ACM Transactions on …, 2019 – dl.acm.org
… Other work has focused on conditioning the scene generation using input from RGB-D frames [Chen et al. 2014], 2D sketches of the scene [Xu et al. 2013], natural language text [Chang et al. 2015; Ma et al … It has been used for text-to-scene generation [Chang et al. 2014] …

Research on 3D Virtual Training Courseware Development System of Civil Aircraft Based on Virtual Reality Technology
Y Tian, M Li, H Liu, S Liu, R Yin – … Artificial Intelligence and Virtual Reality, 2019 – dl.acm.org
… For the component recognition courseware, such as aircraft exterior and cabin roaming, cockpit roaming, etc., the 3D scene function mainly includes requirements on component scene generation, first-view roaming scene, multi-view rendering, model/system multiple realistic …

Learning to Generate Synthetic 3D Training Data through Hybrid Gradient
D Yang, J Deng – arXiv preprint arXiv:1907.00267, 2019 – arxiv.org
… 5.2 NYU Depth Scene perturbation We design our scene generation grammar as an augmentation of collected SUNCG scenes [38] with the cameras from [49] … In Proceedings of the Thirtieth Conference on Uncertainty in Artificial Intelligence, UAI’14, pages 440–448 …

An interactive multi-agent reasoning model for sentiment analysis: a case for computational semiotics
J Akhtar – Artificial Intelligence Review, 2019 – Springer
… on an original student evaluation of teachers dataset, compares the results with deep learning and other baseline techniques, and aims to propose semiotics as a reparative alternative to the dominant dichotomies—rule-based and data-based camps within artificial intelligence …

AI and machine learning: Shaking up the space industr
O Saarela – pdfs.semanticscholar.org
… This t pe of artificial intelligence ( I) has worked well when the inputs to the algorithms fall within the pre-defined mission scope, allowing the pre-built … For satellites, existing images taken during ground processing can likewise be complemented with artificial scene generation …

End-to-End Personalized Humorous Response Generation in Untrimmed Multi-Role Dialogue System
Q Yang, Z He, Z Zhan, R Li, Y Lee, Y Zhang… – IEEE Access, 2019 – ieeexplore.ieee.org
… We estimate the parameters by policy gradient: a probability distribution over actions given an RL state. The RL module system consists of the scene generation agent and the scene judgment audiences. The audiences judge a scene Sc instead of each generated sentence yi …

Web3D-based automatic furniture layout system using recursive case-based reasoning and floor field
P Song, Y Zheng, J Jia, Y Gao – Multimedia Tools and Applications, 2019 – Springer
Furniture layout in a virtual 3D scene is an important and challenging task, as it is time-consuming and requires experience. To address this issue, we propose automatic furniture layout algorithms…

Hypersectral Imaging for Military and Security Applications: Combining Myriad Processing and Sensing Techniques
M Shimoni, R Haelterman… – IEEE Geoscience and …, 2019 – ieeexplore.ieee.org
… These real- time processing platforms were implemented using various computing technologies, including vector processing, field- programmable gate arrays, adaptive (ie, adaptive control of thought), and artificial intelligence …

ROBO: Robust, Fully Neural Object Detection for Robot Soccer
M Szemenyei, V Estivill-Castro – Robot World Cup, 2019 – Springer
… In: Proceedings of the Thirteenth International Conference on Artificial Intelligence and Statistics, pp … Hess, T., Mundt, M., Weis, T., Ramesh, V.: Large-scale stochastic scene generation and semantic annotation for deep convolutional neural network training in the RoboCup SPL …

Give MEANinGS to Robots with Kitchen Clash: A VR Human Computation Serious Game for World Knowledge Accumulation
J Pfau, R Porzel, M Pomarlan, VS Cangalovic… – … Computing and Serious …, 2019 – Springer
… When it comes to undesirable player choices, we will evaluate a knockout system of object alternatives that constrains the variety of choices of the Scene Generation module in … In: Workshops at the Twenty-Ninth AAAI Conference on Artificial Intelligence (2015)Google Scholar. 14 …

Give MEANinGS to Robots with Kitchen Clash: A VR Human Computation Serious Game for World Knowledge Accumulation
S Grudpan, S Höffner, J Bateman… – … Computing and Serious …, 2019 – books.google.com
… we will evaluate a knockout system of object alternatives that constrains the variety of choices of the Scene Generation module in order … In: Workshops at the Twenty-Ninth AAAI Conference on Artificial Intelligence (2015) Pfau, J., Smeddinck, JD, Malaka, R.: Towards deep player …

Structured domain randomization: Bridging the reality gap by context-aware synthetic data
A Prakash, S Boochoon, M Brophy… – … on Robotics and …, 2019 – ieeexplore.ieee.org
… Tsirikoglou et al. [19] introduced procedural modeling for scene generation and also … [30] M. Hodosh, P. Young, and J. Hockenmaier, “Framing image descrip- tion as a ranking task: Data, models and evaluation metrics,” Journal of Artificial Intelligence Research, vol. 47, pp …

Transferring multiscale map styles using generative adversarial networks
Y Kang, S Gao, RE Roth – International Journal of Cartography, 2019 – Taylor & Francis
… for integrating creative, artistic styles into multiscale maps like Google Maps5 and OpenStreetMap (OSM).6 Here, we ask if artificial intelligence (AI) can … & Tuia, 2018; Zhang et al., 2018; Zou, Ni, Zhang, & Wang, 2015; Zhang, Wu, Zhu, & Liu, 2019), scene generation (Deng, Zhu …

A Spoken Dialogue System for Spatial Question Answering in a Physical Blocks World
G Platonov, B Kane, A Gindi, LK Schubert – arXiv preprint arXiv …, 2019 – arxiv.org
… [Chang, Savva, and Manning 2014] Chang, A.; Savva, M.; and Manning, CD 2014. Learning spatial knowledge for text to 3d scene generation. In Proceedings of the 2014 Con- ference on Empirical Methods in Natural Language Pro- cessing (EMNLP), 2028–2038. [Chen et al …

Medical image classification under class imbalance
C Zhang – 2019 – lib.dr.iastate.edu
… Chuanhai Zhang Iowa State University Follow this and additional works at: https://lib.dr.iastate.edu/etd Part of the Artificial Intelligence and Robotics Commons … 4.4. Instrument Scene Generation ….. 52 …

Integrated Energy System Situational Awareness: Concepts, Architecture, and Key Technolo-Gies
R Niu, H Li, X Shang, W Zhang… – 2019 5th International …, 2019 – ieeexplore.ieee.org
… energy characterization, evaluation, prediction, and control method system with comprehensive applicability, which is important for the optimization and regulation of integrated energy systems to realize artificial intelligence … Running scene generation technology …

Dynamic Occlusion Handling for Real-Time AR Applications
J Jorge, RKD Anjos, R Silva – … on Virtual-Reality Continuum and its …, 2019 – dl.acm.org
Page 1. Dynamic Occlusion Handling for Real-Time AR Applications Joaquim Jorge INESC-ID / Técnico / U. Lisboa Lisboa, Portugal jorgej@acm.org Rafael Kuffner dos Anjos INESC-ID / Técnico / U. Lisboa Lisboa, Portugal ranjos@acm.org …

Inference-based creation of synthetic 3D content with ontologies
K Walczak, J Floty?ski – Multimedia Tools and Applications, 2019 – Springer
… Scene generation algorithm. The input data of the scene generation algorithm are the ontology of 3D components, the domain ontology, and the mapping between these two (see Fig. 2). The algorithm consists of the following stages. I … Example of 3D scene generation …

Introspective Environment Modeling
SA Seshia – International Conference on Runtime Verification, 2019 – Springer
… DSN), June 2019Google Scholar. 7. Fremont, DJ, Dreossi, T., Ghosh, S., Yue, X., Sangiovanni-Vincentelli, AL, Seshia, SA: Scenic: a language for scenario specification and scene generation … Seshia, SA, Sadigh, D., Shankar Sastry, S.: Towards Verified Artificial Intelligence …

Task-Oriented Visual Understanding for Scenes and Events
S Qi – 2019 – search.proquest.com
… The param- eters and constraints are automatically learned from the SUNCG [SYZ17a] and 10 ShapeNet [CFG15] datasets. 2. For scene generation, we propose the use of a stochastic grammar model in the form of an attributed Spatial And-Or graph (S-AOG) …

Flood action VR: a virtual reality framework for disaster awareness and emergency response training
Y Sermet, I Demir – ACM SIGGRAPH 2019 Posters, 2019 – dl.acm.org
… Novel devices in virtual reality (VR), and advanced techniques in artificial intelligence (AI), and graphical processer units (GPU) makes it possible for … constructed in ESRI City Engine, and exported in FBX file format which then is imported in Unity3D for use in scene generation …

A new generation of AI: A review and perspective on machine learning technologies applied to smart energy and electric power systems
L Cheng, T Yu – International Journal of Energy Research, 2019 – Wiley Online Library
… Summary. The new generation of artificial intelligence (AI), called AI 2.0, has recently become a research focus. Data?driven AI 2.0 will accelerate the development of smart energy and electric power system (Smart EEPS) … The process of artificial intelligence (AI) evolution …

Transferring Multiscale Map Styles Using Generative Adversarial Networks
Y Kanga, S Gaoa, R Rothb – geods.geography.wisc.edu
… Here, we ask if artificial intelligence (AI) can help illuminate, transfer, and ultimately improve multiscale map styling for cartography, automating some of … classification (Zou et al., 2015; Srivastava et al., 2018; Law et al., 2018; Zhang et al., 2018, 2019), scene generation (Deng et …

Automatic Furniture Layout Based on Functional Area Division
B Yang, L Li, C Song, Z Jiang… – … on Cyberworlds (CW), 2019 – ieeexplore.ieee.org
… [2] studied on hand-drawn sketches that are used to assist in generating indoor scenes for the interactive, diversity scene generation … based 3D object layout using a genetic algorithm[C]// Proceedings of International Conference on Computer Graphics and Artificial Intelligence …

A Comparison of Desktop and Augmented Reality Scenario Based Training Authoring Tools
AV González, S Koh, K Kapalo… – … on Mixed and …, 2019 – ieeexplore.ieee.org
… to generate content. While prior research focuses more on asset creation and object placement, our work explores higher level scene generation with added behaviors under two different interface conditions. AR training applications …

A Programmatic and Semantic Approach to Explaining and DebuggingNeural Network Based Object Detectors
E Kim, D Gopinath, C Pasareanu, S Seshia – arXiv preprint arXiv …, 2019 – arxiv.org
… 2. Background SCENIC [9, 1] is a probabilistic programming language for scenario specification and scene generation. The lan- guage can be used to describe environments for autonomous systems, ie autonomous cars or robots …

Sensor transfer: Learning optimal sensor effect image augmentation for Sim-to-Real domain adaptation
A Carlson, KA Skinner, R Vasudevan… – IEEE Robotics and …, 2019 – ieeexplore.ieee.org
… adversarial network,” arXiv preprint arXiv:1701.05957, 2017. [7] V. Veeravasarapu, C. Rothkopf, and R. Visvanathan, “Adversarially tuned scene generation,” arXiv preprint arXiv:1701.00405, 2017. [8] C. Sakaridis, D. Dai, S. Hecker …

Modeling human intuitions about liquid flow with particle-based simulation
CJ Bates, I Yildirim, JB Tenenbaum… – PLoS computational …, 2019 – journals.plos.org
… escape. Even young children can perceive and interact with liquids in motion (Fig 1B) in ways well beyond the capabilities of modern robots and artificial intelligence (AI) systems. How do people draw rich intuitions about liquids …

Semantic Object Accuracy for Generative Text-to-Image Synthesis
T Hinz, S Heinrich, S Wermter – arXiv preprint arXiv:1910.13321, 2019 – arxiv.org
Page 1. PREPRINT. WORK IN PROGRESS. 1 Semantic Object Accuracy for Generative Text-to-Image Synthesis Tobias Hinz, Stefan Heinrich, and Stefan Wermter Abstract—Generative adversarial networks conditioned on simple …

Text-to-picture tools, systems, and approaches: a survey
J Zakraoui, M Saleh, J Al Ja’am – Multimedia Tools and Applications, 2019 – Springer
… language interface and an interface for visualization purposes requires overcoming profound technical challenges in integrating artificial intelligence techniques, including … It consists of three major processing steps: language processing, KB creation, and scene generation …

Big data analytics for video surveillance
BN Subudhi, DK Rout, A Ghosh – Multimedia Tools and Applications, 2019 – Springer
This article addresses the usage and scope of Big Data Analytics in video surveillance and its potential application areas. The current age of technology provides the users, ample opportunity to…

Generative adversarial networks: A survey and taxonomy
Z Wang, Q She, TE Ward – arXiv preprint arXiv:1906.01529, 2019 – arxiv.org
Page 1. 1 Generative Adversarial Networks: A Survey and Taxonomy Zhengwei Wang, Qi She, Tomás E. Ward Abstract— Generative adversarial networks (GANs) have been extensively studied in the past few years. Arguably …

Grounded Language Processing for Action Understanding and Justification
S Yang – 2019 – search.proquest.com
… To address this problem, recent years have witnessed an increasing interesting on explainable artificial intelligence (XAI) … 48, 66, 96], video sentence alignment [59, 65], scene generation [8], and multi-modal embedding incorporating language and vision [7, 51] …

Desk Organization: Effect of Multimodal Inputs on Spatial Relational Learning
R Rowe, S Singhal, D Yi… – 2019 28th IEEE …, 2019 – ieeexplore.ieee.org
… enable multi-class classification. We describe in more detail how the two forests worked together to perform scene generation as well as how the accuracies of the two forests were computed. For simplicity, the MLN predicate …

Context-based affordance segmentation from 2D images for robot actions
T Lüddecke, T Kulvicius, F Wörgötter – Robotics and Autonomous Systems, 2019 – Elsevier
Skip to main content Skip to article …

A vision of miking: interactive programmatic modeling, sound language composition, and self-learning compilation
D Broman – Proceedings of the 12th ACM SIGPLAN International …, 2019 – dl.acm.org
… 2019. Scenic: a language for scenario specification and scene generation … Church: A Language for Genera- tive Models. In Proceedings of the Twenty-Fourth Conference on Uncer- tainty in Artificial Intelligence (UAI’08). AUAI Press, Arlington, Virginia, United States, 220–229 …

A probabilistic topic model for event-based image classification and multi-label annotation
L Laib, MS Allili, S Ait-Aoudia – Signal Processing: Image Communication, 2019 – Elsevier
Skip to main content Skip to article …

Reaching Out Towards Fully Verified Autonomous Systems
S Sankaranarayanan, S Dutta, S Mover – International Conference on …, 2019 – Springer
… Fremont, DJ, Dreossi, T., Ghosh, S., Yue, X., Sangiovanni-Vincentelli, AL, Seshia, SA: Scenic: a language for scenario specification and scene generation … Hashimoto, DA, Rosman, G., Rus, D., Meireles, O.: Artificial intelligence in surgery: promises and perils. Ann. Surg …

Data Augmentation Based on 3D Model Data for Machine Learning
M Iwasaki, R Yoshioka – 2019 IEEE 4th International …, 2019 – ieeexplore.ieee.org
… images by labels and has long been considered as one of the ultimate goals of artificial intelligence [4]. Image … Vision/Proxy- Virtual-Worlds ) [6] VSR Veeravasarapu, Constantin Rothkopf and Ramesh Visvanathan, “Adversarially Tuned Scene Generation”, arXiv:1701.00405v2 …

Belief Regulated Dual Propagation Nets for Learning Action Effects on Groups of Articulated Objects
AE Tekden, A Erdem, E Erdem, M Imre, MY Seker… – cmpe.boun.edu.tr
… In our simulation experiments, we considered two different configurations for scene generation: a sparse configuration (Fig.2c) where … W. Burgard, “Learning kinematic models for articulated objects,” in Twenty-First International Joint Conference on Artificial Intelligence, 2009 …

ICASSP 2019–2019 IEEE
GB Brighton – 2019 – toc.proceedings.com
… AASP-P1.8: DEEP POLYPHONIC ADSR PIANO NOTE TRANSCRIPTION ….. 246 Rainer Kelz, Sebastian Böck, Austrian Research Institute for Artificial Intelligence, Austria; Gerhard Widmer, Johannes Kepler University, Austria …

What Else Can Fool Deep Learning? Addressing Color Constancy Errors on Deep Neural Network Performance
M Afifi, MS Brown – … of the IEEE International Conference on …, 2019 – openaccess.thecvf.com
Page 1. What Else Can Fool Deep Learning? Addressing Color Constancy Errors on Deep Neural Network Performance Mahmoud Afifi1 1York University, Toronto mafifi@eecs.yorku.ca Michael S Brown1,2 2Samsung AI Center, Toronto mbrown@eecs.yorku.ca Abstract …

Learning neurosymbolic generative models via program synthesis
H Young, O Bastani, M Naik – arXiv preprint arXiv:1901.08565, 2019 – arxiv.org
Page 1. Learning Neurosymbolic Generative Models via Program Synthesis Halley Young 1 Osbert Bastani 1 Mayur Naik 1 Abstract Significant strides have been made toward design- ing better generative models in recent years …

Analytic Continued Fractions for Regression: A Memetic Algorithm Approach
P Moscato, H Sun, MN Haque – arXiv preprint arXiv:2001.00624, 2019 – arxiv.org
… results using analytic continued fractions provides a powerful and interesting new alternative in the quest for compact and interpretable mathematical models for artificial intelligence … forecasting [12], computer game scene generation [13], Boolean classification [14] …

Bi-level multi-time scale scheduling method based on bidding for multi-operator virtual power plant
X Kong, J Xiao, C Wang, K Cui, Q Jin, D Kong – Applied Energy, 2019 – Elsevier
… which wind power providers and electric vehicle aggregators participate in the power market in the form of VPP, and adopted scene generation and reduction … Through the artificial intelligence algorithm, the operator can obtain the bid function coefficients a t , i n + 1 and b t , i n …

Use of Terrain-Based Analysis in Mission Design, Planning and Modeling of Operations of a Lunar Exploration Rover
MS Menon, A Kothandhapani, NS Sundaram… – Space Operations …, 2019 – Springer
… These DTMs are used in conjunction with a tool called Planet and Asteroid Natural Scene Generation Utility (PANGU), developed at University of Dundee, to generate a synthetic scene using the DTM which forms the reference atlas. The resolution of LRO DTM is 5 m/pixel …

Self-supervised damage-avoiding manipulation strategy optimization via mental simulation
T Doernbach – Intelligent Service Robotics, 2019 – Springer
… 2.2 Training scene generation … Since the whole processing circle of training scene generation, mental simulation and manipulation strategy optimization (which includes classifier training) is deferred into load-free times, no delays are inflicted on productive use …

A Survey of Computational Intelligence Techniques for Wind Power Uncertainty Quantification in Smart Grids
H Quan, A Khosravi, D Yang… – IEEE Transactions on …, 2019 – ieeexplore.ieee.org
Page 1. This article has been accepted for inclusion in a future issue of this journal. Content is final as presented, with the exception of pagination. IEEE TRANSACTIONS ON NEURAL NETWORKS AND LEARNING SYSTEMS 1 …

CPS Design with Learning-Enabled Components: A Case Study
C Hartsell, N Mahadevan, S Ramakrishna… – Proceedings of the 30th …, 2019 – dl.acm.org
… Scenic: a language for scenario specification and scene generation. In Proceedings of the 40th ACM SIGPLAN Conference on Programming Language Design and Implementation, pages 63–78. ACM, 2019 … Towards verified artificial intelligence. CoRR, abs/1606.08514, 2016 …

How to build a digital river
RA Brown, GB Pasternack – Earth-Science Reviews, 2019 – Elsevier
Skip to main content Skip to article Elsevier logo: Journals & Books. Create accountSign in. Sign inCreate account. Journals & Books. Help. Download PDFDownload. Share. Export. Advanced.

Virtual Movement from Natural
H Sarma – 2019 – d-nb.info
… The extreme variability in the formation of languages also makes it difficult to understand textual languages [Linell, 2004]. In the field of artificial intelligence, teaching machines how to understand text is extremely im- portant [Kadlec et al., 2016] …

Combining Q-Learning and Search with Amortized Value Estimates
JB Hamrick, V Bapst, A Sanchez-Gonzalez… – arXiv preprint arXiv …, 2019 – arxiv.org
Page 1. Under review as a conference paper at ICLR 2020 COMBINING QLEARNING AND SEARCH WITH AMORTIZED VALUE ESTIMATES Jessica B. Hamrick DeepMind jhamrick@google.com Victor Bapst DeepMind vbapst@google.com …

Novel Advances in the Development of Machine Learning Solutions for Scientific Programming
V García-Díaz, CE Montenegro-Marin – downloads.hindawi.com
Page 1. Scientific Programming Novel Advances in the Development of Machine Learning Solutions for Scientific Programming Lead Guest Editor: Vicente García-Díaz Guest Editors: Edward R. Núñez-Valdez, Vijender K. Solanki, and Carlos Enrique Montenegro-Marin Page 2 …

27markerless motion capture for 3D human model animation using depth camera
M Galinium, J Yapri, J Purnama – Telkomnika, 2019 – researchgate.net
… International Journal of Advanced Research in Artificial Intelligence. 2013: 2(7): 24-28. [12] Li C, Yin C, Lu J, Ma L. Automatic 3D Scene Generation based on Maya. Proceedings of 10th International Conference on Computer-Aided Industrial Design & Conceptual Design …

Position paper on the challenges posed by modern applications to cyber-physical systems theory
F Allgöwer, JB de Sousa, J Kapinski… – Nonlinear Analysis …, 2019 – Elsevier
… in Fig. 1. The first framework shown in Fig. 1(a) is that of the levels of mental activities humans are capable of, as outlined by Marvin Minsky in the context of artificial intelligence (AI) [1]. The second framework shown in Fig. 1(b …

TextInContext: On the Way to a Framework for Measuring the Context-Sensitive Complexity of Educationally Relevant Texts—A Combined Cognitive and …
A Mehler, V Ramesh – Frontiers and Advances in Positive Learning in the …, 2019 – Springer
We develop a framework for modeling the context sensitivity of text interpretation. As a point of reference, we focus on the complexity of educational texts. To open up a broader basis for…

Analytics for awareness in maritime surveillance: from data to tactical insight
T Caelli, J Mukerjee, E Sparks – The Journal of Defense …, 2019 – journals.sagepub.com
Although significant effort has occurred into developing realistic simulation environments for maritime surveillance, relatively little attention has been given…

A Novel Multi-View Table Tennis Umpiring Framework
H Myint – 2019 – oro.open.ac.uk
Page 1. Open Research Online The Open University’s repository of research publications and other research outputs A Novel Multi-View Table Tennis Umpiring Framework Thesis How to cite: Myint, Hnin (2019). A Novel Multi-View Table Tennis Umpiring Framework …

Visual Perception for Robotic Spatial Understanding
J Owens – 2019 – search.proquest.com
… Department of Computer and Information Science Jianbo Shi, Professor, Department of Computer and Information Science Jean Gallier, Professor, Department of Computer and Information Science Nicholas Roy, Professor, Computer Science and Artificial Intelligence Lab, MIT …

Smart Technologies for Unmanned Surface Vessels
US Vessels – pdfs.semanticscholar.org
… larger volumes have been produced. Artificial Intelligence (AI) algorithms such as Machine Learning (ML), have evolved, and so have the Graph- ical Processing Units (GPU) that process the ML algorithms. AI methods are often …