Scene Understanding & Natural Language 2016


Resources:

Wikipedia:

See also:

100 Best GraphLab VideosText-to-Image Systems


Natural Language Communication with Robots.
Y Bisk, D Yuret, D Marcu – HLT-NAACL, 2016 – pdfs.semanticscholar.org
… cO2016 Association for Computational Linguistics Natural Language Communication with Robots … We show how one can collect meaningful training data and we propose three neural architectures for interpreting contextually grounded natural language commands. …

Automatic Description Generation from Images: A Survey of Models, Datasets, and Evaluation Measures.
R Bernardi, R Cakici, D Elliott, A Erdem, E Erdem… – J. Artif. Intell. Res …, 2016 – jair.org
… 1. Introduction Over the past two decades, the fields of natural language processing (NLP) and computer vision (CV) have seen great advances in their respective goals of analyzing and generating text, and of understanding images and videos. …

Situated Language Understanding with Human-like and Visualization-Based Transparency.
L Perlmutter, E Kernfeld… – Robotics: Science and …, 2016 – roboticsproceedings.org
… Fig. 1: LUCIT combines speech recognition, pointing detection, and scene understanding to interpret situated natural language commands. LUCIT’s transparency mechanisms allow the user view to the output of these intermediate internal processes. to confirm it. …

Relational grounded language learning
L Becerra-Bonache, H Blockeel… – Proceedings of the …, 2016 – lirias.kuleuven.be
… Conf. on Computational Natural Language Learning, pp. … IEEE, (2015). [16] CL Zitnick, R. Vedantam, and D. Parikh, ‘Adopting abstract images for semantic scene understanding’, IEEE Transactions on Pattern Anal- ysis and Machine Intelligence, 38(4), 627–638, (2016).

What Is Where: Inferring Containment Relations from Videos.
W Liang, Y Zhao, Y Zhu, SC Zhu – IJCAI, 2016 – stat.ucla.edu
… 1 Introduction and Motivations For many AI tasks, such as scene understanding in visual perception, task planning in robot autonomy, and symbol grounding in natural language understanding, a key problem is to infer “what is where over time”. …

RGB-D scene labeling with long short-term memorized fusion model
Z Li, Y Gan, X Liang, Y Yu… – arXiv preprint arXiv …, 2016 – pdfs.semanticscholar.org
… ac- cess of affordable depth sensor, scene labeling in RGB-D images [14,22,23,12,13,24] enables a rapid progress of scene understanding. … there exists some problems for applying RNN on image processing due to the fact that the in- put in natural language processing (NLP …

Modeling commonsense reasoning via analogical chaining: A preliminary report
J Blass, K Forbus – Proceedings of CogSci, 2016 – mindmodeling.org
… First, we plan to expand our NLU capabilities to support fully automatic construction of CSUs from natural language, rather than mixing automatic generation with some manual editing, both to reduce tailorability and to … Simulation as an engine of physical scene understanding. …

Deepiu: an architecture for image understanding
Y Yang, UMD EDU, Y Aloimonos… – Adv Cogn Syst Google …, 2016 – researchgate.net
… Similarly, “understanding” in an automated environment can be tested by asking questions and an intelligent system attempting to “understand” any concept should have the ability to answer them. Natural Language Understanding systems (Katz et al. (2001),Weston et al. …

Multimodal Person Discovery in Broadcast TV at MediaEval 2016.
H Bredin, C Barras, C Guinaudeau – MediaEval, 2016 – ceur-ws.org
… 4. BASELINE AND METADATA This task targets researchers from several communities in- cluding multimedia, computer vision, speech and natural language processing. … Scene understanding for identifying persons in TV shows: beyond face authentication. In CBMI, 2014. …

PNN Based Character Recognition in Natural Scene Images
SV Seeri, JD Pujari, PS Hiremath – researchgate.net
… I. INTRODUCTION EXT information in a scene image is the key clue for scene understanding. … Text consists of set of words which are formed by sequence of characters of varying size and styles. It is hard to develop a common text descriptor for any natural language. …

Data Science News
KAI Legal – pdfs.semanticscholar.org
… Kyunghyun Cho Talks Image Caption Generation NYU Center for Data Science from March 28, 2016 Kyunghyun Cho is an Assistant Professor at NYU’s Center for Data Science, and conducts research in the field of natural language processing. …

Geometric Concept Acquisition by Deep Reinforcement Learning
A Kuefler – stanford.edu
… Understanding natural language. Cognitive psychology, 3(1):1–191, 1972. … A comparative evaluation of approximate probabilistic simulation and deep neural networks as accounts of human physical scene understanding. arXiv preprint arXiv:1605.01138, 2016. …

A Proposal for Common Dataset in Neural-Symbolic Reasoning Studies.
Ö Yilmaz, ASA Garcez, DL Silver – NeSy@ HLAI, 2016 – pdfs.semanticscholar.org
… and reasoning are limited to image description text that is unstructured, and not amenable to traditional natural language process- ing … Overall the dataset enables a wide range of scene understanding applications, which typically require high level symbol manipulation and …

Review of deep learning approaches for image caption generation
AS Sisodiya – 2016 – iitk.ac.in
… To capture the correlation between two modalities ie visual and natural language we need to map both these to some same space so at learn the … Image caption generation is a core part of scene understanding which is one of the primary goals of vision right from the start. …

Research proposal: Evaluating multi-modal deep learning systems with micro-worlds
A Kuhnle – 2016 – cl.cam.ac.uk
… forthcoming. Multi-agent cooperation and the emergence of (natural) language. International Conference on Learning Representations (ICLR 2017) (under review). … 2016. Adopting abstract images for semantic scene understanding. …

Personal Statement
J Košecká – cs.gmu.edu
… Based Localization Using Robust and Efficient Techniques, F. Li, PhD 2007, Strangeness based Feature Selection and G. Singh, PhD 2014, Visual Scene Understanding though Semantic … This work combined methods used in Natural Language Processing and Computer Vision …

Improving Semantic Video Segmentation by Dynamic Scene Integration
FG Zanjani, M van Gerven – ru.nl
… Introduction Low-level video segmentation is an important objective in many application areas such as robotics, object tracking, video coding, video perception, action recognition and scene understanding. … The cityscapes dataset for semantic urban scene understanding. …

Connecting Images and Natural Language
A Karpathy – 2016 – pdfs.semanticscholar.org
… Humans find it easy to accomplish a wide variety of tasks that involve complex visual recognition and scene understanding, tasks that involve communication in natural language and tasks that combine translation between the two modalities. …

Learning to Talk about Events Grounding Language Acquisition in Intuitive Theories and Event Cognition
E Wittenberg, M Kline, JK Hartshorne – pdfs.semanticscholar.org
… Battaglia, P., Hamrick, J., & Tenenbaum, JB (2013). Sim- ulation as an engine of physical scene understanding. … In K. von Heusinger, C. Maienborn, & P. Portner (Eds.), Semantics: An international handbook of natural language meaning I. Berlin: Mouton de Gruyter. …

Commonsense for Making Sense of Data.
SN Chowdhury – PhD@ VLDB, 2016 – pdfs.semanticscholar.org
… NLP and Computer Vision: Existing research on auto- matic image annotations [28], description generation [27, 16, 14], scene understanding [5], and image extraction … natural language queries [13] point towards the ongoing col- laboration of the NLP and CV communities. …

Commonsense for Making Sense of Data
S Nag Chowdhury – VLDB 2016 PhD Workshop, 2016 – pubman.mpdl.mpg.de
… NLP and Computer Vision: Existing research on auto- matic image annotations [28], description generation [27, 16, 14], scene understanding [5], and image extraction … natural language queries [13] point towards the ongoing col- laboration of the NLP and CV communities. …

Computer Vision and Deep Learning for Automated Surveillance Technology
T de Planque – stanford.edu
… Moreover, the above semantic knowledge has to be ex- pressed in a natural language like English, which means that a language model is needed in addition to visual understand- ing. … ”Towards Total Scene Understanding: Classification, Annotation and Segmentation in an …

Computer Vision Lab, University of Maryland, College Park, MD 20740, USA.{cxy, yzyang, fer, yiannis}@ umiacs. umd. edu* henryzhao4321@ gmail. com
C Ye, C Zhao, Y Yang, C Fermüller, Y Aloimonos – pdfs.semanticscholar.org
… Different applications in computer vision, natural language processing and robotics are demonstrated as experiments. … Categories and Subject Descriptors D.0 [Software]: General; I.2.10 [Artificial Intelligence]: Vision and Scene Understanding …

A FRAMEWORK FOR STRING TRANSFORMATION USING PROBABILISTIC APPROACH
M PRIYA, T SRISUPRIYA – ijarcsa.org
… String transformation can be formalized as the part of natural language processing, data mining and information retrieval. … the structures to be learned become more complex than the amount of training data (eg, in machine translation, scene understanding, biological process …

Applications of Various Artificial Intelligence Techniques in Software Engineering
D Saini – ijrest.net
… computer systems that exhibit some form of intelligence and attempts to apply such knowledge to the design of computer based systems that can understand a natural language or … 28 context awareness, scene understanding (Aarts EHL, 2003), …

Know2Look: Commonsense Knowledge for Visual Search.
SN Chowdhury, N Tandon… – AKBC@ NAACL …, 2016 – pdfs.semanticscholar.org
… 58 Page 3. 2014), description generation (Vinyals et al., 2014; Ordonez et al., 2011; Mitchell et al., 2012), scene understanding (Farhadi et al., 2010), image retrieval through natural language queries (Malinowski and Fritz, 2014) etc. …

Recognition of Sign and Text Using LVQ and SVM
PVN Kumar, R Ganesamoorthy – irjaes.com
… The term applies both to the mental processes used by humans, and to artificial processes of natural language processing. … In more global context, it can contribute to the scene understanding of traffic context (eg, if the car is driving in a city or on a freeway). …

School of Computing
R Cruise – 2016 – pdfs.semanticscholar.org
… Traditionally, CV approaches the problem of scene understanding as one of fi nd- ing methods to transform between input images … realisation that much cognitive representation and processing of spatial data is qualitative – eg most everyday natural language spatial expressions …

Lantern: A Query Language for Visual Concept Retrieval
W Crichton – cs.cmu.edu
… Just as advances in relation databases and natural language processing have enabled text to be analyzed en masse, so too do we need tools for handling the enormous quantity of visual data to answer these kinds of questions. …

A Novel Approach for On-road Vehicle Detection and Tracking
I El Jaafari, M El Ansari, L Koutti… – INTERNATIONAL …, 2016 – researchgate.net
… is a required critical and important task not only for ADAS, but also for other real-world applications including urban scene understanding, automated driving … Sliding windows have also been applied within the field of natural language processing for collocation detection [38]. …

Review of state-of-the-arts in artificial intelligence. Present and future of AI.
V Shakirov – alpha.sinp.msu.ru
… The same holds true for LSTMs in translation, natural language generation etc. … ”Attend, Infer, Repeat: Fast Scene Understanding with Generative Models” http: //arxiv.org/abs/ 1603.08575 [48] Alec Radford, Luke Metz, Soumith Chintala. …

What is Orientation?
J Krukar, A Schwering – 13th Biannual Conference of the … – pdfs.semanticscholar.org
… recognition. The topics of interest address various is- sues related to sketching and sketch understanding, as well as gestures, scene understanding and interpretation, and the generation of image schemas to de- pict concepts. …

Blending learning and inference in conditional random fields
T Hazan, AG Schwing, R Urtasun – Journal of Machine Learning Research, 2016 – jmlr.org
… and inference of structured models drives much of the research in machine learning applications, from computer vision and natural language processing to … was shown to improve the state-of-the-art in various computer vision tasks, including 2D scene understanding (Yao et al …

Visual Genome
R Krishna, Y Zhu, O Groth, J Johnson, K Hata, J Kravitz… – pdfs.semanticscholar.org
… that underpin today’s advances in computational visual understanding. While the progress is exciting, we are still far from reaching the goal of comprehensive scene understanding. As Fig- ure 1 shows, existing models would …

Interactive Consensus Agreement Games For Labeling Images
P Upchurch, D Sedra, A Mullen, H Hirsh, K Bala – 2016 – dmsedra.com
… Science, Cornell University Abstract Scene understanding algorithms in computer vision are im- proving dramatically by training deep convolutional neural networks on millions of accurately annotated images. Col- lecting large …

Interactive spatiotemporal cognition: Data, theories, architectures, and autonomy
S Khemlani, JG Trafton – Psychological Bulletin – pdfs.semanticscholar.org
… Simulation as an engine of physical scene understanding. Proceedings of the National Academy of Sciences, 110, 18327-18332. Bonato, M., Zorzi, M., & Umiltà, C. (2012). … Huffman, SB, & Laird, JE (2014). Learning procedures from interactive natural language instructions. …

The QuEST for multi-sensor big data ISR situation understanding
J Patrick, E Blaschc, J Trumpfhellerd – Proc. of SPIE Vol, 2016 – researchgate.net
… There have been considerable advances in what is often described as Iscene understanding.” Scene understanding is the ability to understand and describe a scene or picture in natural language. The recent advances have …

EFFICIENT SEMANTIC SEGMENTATION OF MAN-MADE SCENES USING FULLY-CONNECTED CONDITIONAL RANDOM FIELD.
W Li, M Ying Yang – International Archives of the …, 2016 – … -remote-sens-spatial-inf-sci.net
… Convolutional patch networks with spatial prior for road detection and urban scene understanding. In: International Con- ference on Computer Vision Theory and Applications (VISAPP), pp. 510–517. … Parsing natural scenes and natural language with recursive neural networks. …

Automatic Indoor 3D Scene Classification using RGB-D data
N de Wolf – 2016 – esc.fnwi.uva.nl
… D data [23]. In order to make optimal use of the depth data that is generated from the depth sensors, the scope of the scene understanding tasks will be limited to indoor Page 8. Introduction 3 … of natural language processing, that works by representing a document as a vector that …

The meaning of action: A review on action recognition and mapping
D Kragic, CW Geib – academia.edu
… An application of great potentials are 1 Page 3. automatic scene understanding systems that include the interpretation of the observed actions such as what actions are executed, where they are executed, who is involved, and even a prediction of what the …

THE NATIONAL ARTIFICIAL INTELLIGENCE RESEARCH AND DEVELOPMENT STRATEGIC PLAN
S PLAN – 2016 – raincent.com
… for humans, and improve themselves”.2 This historic meeting set the stage for decades of government and industry research in AI, including advances in perception, automated reasoning/planning, cognitive systems, machine learning, natural language processing, robotics, …

Conference 9842: Signal Processing, Sensor/Information Fusion, and Target Recognition XXV
B Balaji, R Sithiravel, A Damini – TECHNICAL SUMMARIES• – spie.org
… MIDAT combines natural language processing, multi-hypothesis tracking, and Multi-INT Activity Pattern Learning and Exploitation (MAPLE) technologies in an end-to-end lab prototype that processes textual products produced by video analysts, infers POLs, and highlights …

Guaranteed Parameter Estimation of Discrete Energy Minimization for 3D Scene Parsing
M Li – 2016 – pdfs.semanticscholar.org
… 1 [34], is a popular model for many problems in com- puter vision, machine learning, bioinformatics, and natural language processing. … vision practice, energy minimization has found its place in semantic segmentation [69], pose estimation [102], scene understanding [76], depth …

Photographic credits Céline Ribordy, Sion? Sedrik Nemeth, Sion? Nicolas Sedlatchek, Sion? Idiap, Martigny Graphic Design Formaz—Andenmatten, Sion …
PM Fellay – pdfs.semanticscholar.org
… Natural Language Processing Andrei Popescu-Belis … The Natural Language Processing group studies how the seman- tic and pragmatic analysis of texts can improve the execution of two important tasks—machine translation and information retrieval. …

Neural Network Architectures for Reverberated Lecture Speech Recognition
M Ritter – 2016 – isl.anthropomatik.kit.edu
… This makes them ideal for machines learning. During the last decade they have outperformed many other models in various machine learning tasks, including speech recognition, natural language processing and image classification [11, 12, 13]. …

Making Sense of Sensors: End-to-End Algorithms and Infrastructure Design from Wearable-Devices to Data Centers
O Tickoo, R Iyer – 2016 – Springer
… 30 2.3.3 Automatic Speech Recognition (ASR) …. 30 2.3.4 Natural Language Processing (NLP) …. 32 Page 6. ? CONTENTS vi … 64 4.7 Concluding Thoughts on Scene Understanding….. 65 …

On Machine Perception of Sound
A Kumar – 2016 – pdfs.semanticscholar.org
Page 1. November 20, 2016 DRAFT On Machine Perception of Sound Ph.D. Thesis Proposal Anurag Kumar Language Technologies Institute School of Computer Science Carnegie Mellon University Pittsburgh, PA 15213 Thesis …

Unsupervised Structured Learning Of Human Activities For Robot Perception
C Wu – 2016 – www-cs.stanford.edu
… niques are unsupervised and structured modeled, they are easily extended and scaled to other areas, such as natural language processing, robotic plan- ning/manipulation or multimedia analysis. … Scene understanding. Scene understanding from 2D images has been widely …

A Research about Cloud Reliability Evaluation Method.
Y Qin – International Journal of Simulation–Systems, Science …, 2016 – ijssst.info
Page 1. YI QIN: A RESEARCH ABOUT CLOUD RELIABILITY EVALUATION METHOD DOI 10.5013/IJSSST.a.17.31.12 12.1 ISSN: 1473-804x online, 1473-8031 print A Research about Cloud Reliability Evaluation Method Yi Qin*1 …

3.35 Open Problems and State-of-Art of Session Types
N Yoshida – Compositional Verification Methods for Next …, 2016 – pdfs.semanticscholar.org
… dagstuhl. de/15192 1998 ACM Subject Classification I. 2.10 Visual and Scene Understanding; I. 4.8 Scene Analysis; J. 5 Arts and Humanities. Keywords and phrases Cast Shadows, Perception, Computer Vision, Space Cognition Digital Object Identifier 10.4230/DagRep. 5.5. …

MASTERS OF SCIENCE
R Krishna – 2016 – pdfs.semanticscholar.org
… Together, we show how semantic image search can be improved by converting natural language queries into scene graphs and then mapping them to images. … While the progress is exciting, we are still far from reaching the goal of comprehensive scene understanding. …

The computational origin of representation and conceptual change
ST Piantadosi – 2016 – colala.bcs.rochester.edu
… expressive power (Hindley & Seldin, 1986). In cognitive research, combinatory logic is primarily seen in formal theories of natural language semantics (Steedman, 2001; Jacobson, 1999). However, its general usefulness is demonstrated …

Simultaneous localization and mapping: Present, future, and the robust-perception age
C Cadena, L Carlone, H Carrillo… – CoRR, vol. abs …, 2016 – pdfs.semanticscholar.org
Page 1. 1 Simultaneous Localization And Mapping: Present, Future, and the Robust-Perception Age Cesar Cadena, Luca Carlone, Henry Carrillo, Yasir Latif, Davide Scaramuzza, José Neira, Ian D. Reid, John J. Leonard Abstract …

Learning Contextualized Semantics from Co-occurring Terms via a Siamese Architecture
KC Ubai Sandouk – research.manchester.ac.uk
… among terms. Latent Dirichlet Allocation (LDA) (Blei, Ng, & Jordan, 2003) and Probabilistic Latent Semantic Analysis (PLSA) (Hofmann, 1999) are the most prominent topic models used in text and natural language processing. In …

IMAGE CAPTIONING WITH RECURRENT NEURAL NETWORKS
BJ KVITA – dspace.vutbr.cz
… practical than shallow ones. During this reinvention, neural nets have been successfully applied in multiple fields like computer vision [21], speech recognition [18], and natural language modeling [40]. Nowadays, various useful …

Mind Genomics: A Guide to Data-Driven Marketing Strategy
V Milutinovic, J Salom – 2016 – Springer
Page 1. 123 SPRINGER BRIEFS IN BUSINESS Veljko Milutinovic Jakob Salom Mind Genomics A Guide to Data- Driven Marketing Strategy Page 2. SpringerBriefs in Business Page 3. More information about this series at http://www.springer.com/series/8860 Page 4. …

Fine-grained Recognition: Data, Recognition, and Application
J Krause – 2016 – pdfs.semanticscholar.org
… more detail, fine-grained recognition can be used for improved scene understanding, studying society, and even analyzing biodiversity. … from images. Finally, I present a work looking at finer-grained natural language descriptions of images. iv Page 3. Acknowledgments …

Event Recognition and Forecasting Tech-nology
E Michelioudakis, A Skarlatidis, E Alevizos, A Artikis… – 2016 – speedd-project.eu
… Since this field is relatively new, without a substantial number of contributions coming from researchers directly involved with CER, we have chosen to adopt a broader perspective and include methods targeting activity recognition and scene understanding on image sequences …

Active Learning for High Dimensional Inputs using Bayesian Convolutional Neural Networks
R Islam – 2016 – mlsalt.eng.cam.ac.uk
Page 1. Active Learning for High Dimensional Inputs using Bayesian Convolutional Neural Networks Riashat Islam Department of Engineering University of Cambridge M.Phil in Machine Learning, Speech and Language Technology …