Scene Understanding & Natural Language 2013


Resources:

See also:

100 Best GraphLab VideosText-to-Image Systems


Jointly Learning to Parse and Perceive: Connecting Natural Language to the Physical World. J Krishnamurthy, T Kollar – TACL, 2013 – jayantk.com … Figure 1: LSP applied to scene understanding. Given an environment containing a set of objects (left), and a natural language query, LSP produces a semantic parse, logical knowledge base, grounding and denotation (middle), using only language/denotation pairs (right) for … Cited by 18 Related articles All 11 versions

Incrementally biasing visual search using natural language input E Krause, R Cantrell, E Potapova, M Zillich… – Proceedings of the …, 2013 – dl.acm.org … Categories and Subject Descriptors I.2.7 [Artificial Intelligence]: Natural Language Processing— Discourse, Language parsing and understanding; I.2.10 [Vision and Scene Understanding]: 3D/stereo scene analysis; I.4.8 [Image Processing and Computer Vision]: Scene Analysis … Cited by 4 Related articles All 5 versions

Learning hierarchical features for scene labeling C Farabet, C Couprie, L Najman… – Pattern Analysis and …, 2013 – ieeexplore.ieee.org … This suggests that recognition architectures for vision (and for other modalities such as audio and natural language) … While ConvNets have been used successfully for a number of image labeling problems, image-level tasks such as full-scene understanding (pixelwise labeling … Cited by 96 Related articles All 30 versions

Bringing semantics into focus using visual abstraction CL Zitnick, D Parikh – Computer Vision and Pattern …, 2013 – ieeexplore.ieee.org … Works in learning semantic attributes [2, 6, 25] are becoming popular for enabling hu- mans and machines to communicate using natural language. … For instance, the study of abstract scenes may help research in semantic scene understanding in real images by suggesting to … Cited by 9 Related articles All 16 versions

Basic level scene understanding: categories, attributes and structures J Xiao, J Hays, BC Russell, G Patterson… – Frontiers in …, 2013 – ncbi.nlm.nih.gov … Figure 10 Figure 10. Unified 3D scene parsing for basic level scene understanding. … Inspired by natural language parsing (NLP) using structural SVM (Joachims et al., 2009), we encode both types of rules—image likelihood and context constraints—into the feature vector f(x,y … Cited by 3 Related articles All 17 versions

ImageSpirit: Verbal guided image parsing MM Cheng, S Zheng, WY Lin, J Warrell… – arXiv preprint arXiv: …, 2013 – arxiv.org … However, most speech interface research is focused on natural language processing and to our knowledge there has been no prior work addressing image region selection through speech. … Our paper’s focus is not on natural language interpretation. … Cited by 4 Related articles All 6 versions

A markov logic framework for recognizing complex events from multimodal data YC Song, H Kautz, J Allen, M Swift, Y Li, J Luo… – Proceedings of the 15th …, 2013 – dl.acm.org … Descriptors I.2.10 [Vision and Scene Understanding]: 3D/stereo scene analysis; I.2.4 [Knowledge Representation For- malisms and Methods]: Predicate logic; I.2.3 [Deduction and Theorem Proving]: Probabilistic reasoning; I.2.7 [Natural Language Processing]: Language … Cited by 2 Related articles All 13 versions

Toward Interactive Grounded Language Acqusition. T Kollar, J Krishnamurthy, G Strimel – Robotics: Science and …, 2013 – andrew.cmu.edu … INTERACTIVE TRAINING This section describes the interactive scene understanding system that we constructed for our experiments … mode enables a user to generate training examples for LSP, while the generation mode allows a user to obtain natural language descriptions of … Cited by 1 Related articles All 10 versions

Learning the visual interpretation of sentences CL Zitnick, D Parikh… – Computer Vision (ICCV), …, 2013 – ieeexplore.ieee.org … As the fields of Natural Language Processing (NLP) and computer vision progress we may move towards more descriptive and intuitive sentence-based … This allows us to focus on the core problem of se- mantic scene understanding, while avoiding problems that arise with the … Cited by 9 Related articles All 11 versions

A sentence is worth a thousand pixels S Fidler, A Sharma, R Urtasun – Computer Vision and Pattern …, 2013 – ieeexplore.ieee.org … Abstract We are interested in holistic scene understanding where images are accompanied with text in the form of complex sentential descriptions. … Here we are interested in exploiting textual information for semantic scene understanding. … Cited by 6 Related articles All 14 versions

Fusion of spatially referring natural language statements with random set theoretic likelihoods AN Bishop, B Ristic – Aerospace and Electronic Systems, IEEE …, 2013 – ieeexplore.ieee.org … Natural Language Statements … This work concerns an automatic information fusion scheme for state estimation where the inputs (or measurements), used to reduce the uncertainty in the state of a subject, are in the form of natural language propositions. … Cited by 1 Related articles All 3 versions

Attribute dominance: What pops out? N Turakhia, D Parikh – Computer Vision (ICCV), 2013 IEEE …, 2013 – ieeexplore.ieee.org … Attributes have been used to learn and evaluate models of deeper scene understanding [8, 20] that reason about properties of ob- jects … Efforts at generating natural language sentences can benefit from our work that determines which attributes ought to be mentioned in the first … Cited by 7 Related articles All 5 versions

Sonification of images for the visually impaired using a multi-level approach M Banf, V Blanz – Proceedings of the 4th Augmented Human …, 2013 – dl.acm.org … Figure 1: A blind participant during image explo- ration using our software and a Touch Screen. is then output in natural language on a speech synthesis de- vice. … Geared to scene understanding, rather than object recog- nition by the user. … Cited by 4 Related articles All 2 versions

A maximum entropy based data-driven 3D scene generation MA DEMA, H Sari-Sarraf – International Journal of Semantic …, 2013 – World Scientific … of these resources is making it easier for researchers to participate in study areas that focus on scene understanding, scene generation … be men- tioned that MaxEnt model has been used extensively in the past for diÆerent applications such as natural language processing [11 … Cited by 2 Related articles

Kernel-based discriminative re-ranking for spoken command understanding in hri R Basili, E Bastianelli, G Castellucci, D Nardi… – AI* IA 2013: Advances in …, 2013 – Springer … These represent a scene understanding paradigm, that in turn corresponds to the robot’s executable plan, thus providing an integrated model for language … In Natural Language Processing, re-ranking has been largely used such as in parse tree re-ranking [3] to improve the … Cited by 2 Related articles All 2 versions

Recursive neural networks based on PSO for image parsing GR Cai, SL Chen – Abstract and Applied Analysis, 2013 – hindawi.com … neighboring segments. The graphical model inference pulls out the most consistent set of segments which covers the image [1]. Recently, these ideas have been combined to provide more detailed scene understanding [12–15]. It … Cited by 1 Related articles All 9 versions

Study of Various Video Annotation Techniques K Khurana, MB Chandak – … Journal of Advanced Research in Computer …, 2013 – ijarcce.com … LREC Workshop on Annotation Standards for Temporal Information in Natural Language, pp. 36-43, 2002. … [32] R. Brehar, C. Fortuna, et al., “Spatio-temporal Reasoning for Traffic Scene Understanding”, Intelligent Computer Communication and Processing IEEE, pp. … Cited by 5 Related articles All 2 versions

Sentribute: image sentiment analysis from a mid-level perspective J Yuan, S Mcdonough, Q You, J Luo – Proceedings of the Second …, 2013 – dl.acm.org … using SUN Database2, the first large- scale scene attribute database, initially designed for high- level scene understanding and fine … In Proceedings of the conference on Human Language Technology and Empirical Methods in Natural Language Processing, pages 347–354. … Cited by 2 Related articles All 3 versions

Balance between object and background: Object-enhanced features for scene image classification Z Ji, J Wang, Y Su, Z Song, S Xing – Neurocomputing, 2013 – Elsevier … most popular one. The idea of BOW is inspired by the success of textual words in natural language processing area, where documents are represented by a bag of textual words from a vocabulary dictionary. Analogously, an … Cited by 3 Related articles All 2 versions

Visual-Semantic Scene Understanding by Sharing Labels in a Context Network I Chakraborty, A Elgammal – arXiv preprint arXiv:1309.3809, 2013 – arxiv.org … To facilitate this joint inference, we conceptualize scene understanding as a top-down integration of lexical and visual spaces via object names. … One strand of research is aimed at generating natural language sen- tences from objects and their inter-relations [7]. Our prob- lem is … Related articles All 2 versions

Efficient approach for scene understanding by low confidence region boosting I Shopovska, Z Ivanovski – Multimedia Signal Processing ( …, 2013 – ieeexplore.ieee.org … pp. 57 – 70. [24] Socher, Richard, et al., “Parsing natural scenes and natural language with recursive neural networks.” 2011. ICML. [25] Pawan Kumar, M. and Koller, Daphne., “Efficiently selecting regions for scene understanding.” 2010. CVPR. … Related articles

Dynamic scene understanding by improved sparse topical coding W Fu, J Wang, H Lu, S Ma – Pattern Recognition, 2013 – Elsevier … Pattern Recognition. Volume 46, Issue 7, July 2013, Pages 1841–1850. Cover image Cover image. Dynamic scene understanding by improved sparse topical coding. … Highlights. ? Dynamic scene understanding is addressed as a sparse topical coding problem. … Related articles All 6 versions

Understanding vehicular traffic behavior from video: a survey of unsupervised approaches BT Morris, MM Trivedi – Journal of …, 2013 – electronicimaging.spiedigitallibrary. … 1. Cited by 2 Related articles All 9 versions

Ursus: A Robotic Assistant for Training of Children with Motor Impairments CS Mejías, C Echevarría, P Nuñez, L Manso… – Converging Clinical and …, 2013 – Springer … information. According to the natural language (ie, robot speech), Verbio Text- to-Speech system has been used. Besides … for RoboComp. In: Workshop on Recognition and Action for Scene Understanding Reacts, Málaga (2011) Related articles All 2 versions

Automatic Recognition and Understanding of the Driving Environment for Driver Feedback J Zhou, LE Navarro-Serment, M Hebert – utc.ices.cmu.edu … 3265–3272. [4] BERGER, AL, PIETRA, SAD, AND PIETRA, VJD A maximum entropy approach to natural language processing. … 2823 –2830. [11] GEIGER, A., LAUER, M., AND URTASUN, R. A generative model for 3d urban scene understanding from movable platforms. … Related articles All 2 versions

Parsing Natural Scenes Using Recursive Neural Network S Gupta, V Mishra – 2013 – cse.iitk.ac.in … Page 2. 2. PREVIOUS WORK Scene understanding is one of the most fascinating task in AI. … [1] Richard Socher and Cliff C. Lin and Andrew Y. Ng and Christopher D. Manning, Parsing Natural Scenes and Natural Language with Recursive Neural Networks, ICML 2011 … Related articles All 3 versions

ViCoMo: visual context modeling for scene understanding in video surveillance IM Creusen, S Javanbakhti… – Journal of …, 2013 – proceedings.spiedigitallibrary.org 1. Cited by 1 Related articles All 6 versions

Towards Ontological Cognitive System C Fernandez, J Gonzàlez, JMRS Tavares… – Topics in Medical Image …, 2013 – Springer … 2005–2007): It seeks to work out a new structure contributing to an increase in the evaluation of video scene understanding. … will comprise an internal logic-based representation, which enables it to comment each individual subset, giving natural language explanations of why … Related articles All 7 versions

Fusing video and text data by integrating appearance and behavior similarity G Levchuk, C Shabarekh – SPIE Defense, …, 2013 – proceedings.spiedigitallibrary.org … Co-reference resolution, multi-modal fusion, computer vision, activity recognition, behavior signatures, video-to-text, adversarial reasoning, scene understanding … Entity co-reference resolution is a well-studied problem in natural language processing (Ponzetto, and Strube, 2006 … Related articles All 3 versions

Towards Cognitive Image Interpretation-Qualitative Descriptors, Domain Knowledge and Narrative Generation. Z Falomir – CCIA, 2013 – books.google.com … qualitative descriptors, natural language, domain knowledge, object detectors 1. Introduction Companion robots and ambient intelligent systems need to … However, although many advances have been carried out in computer vision, scene understanding is still an on-going area … Related articles All 4 versions

Multi-party interaction with a virtual character and a human-like robot Z Yumak, N Magnenat-Thalmann – … of the 19th ACM Symposium on …, 2013 – dl.acm.org … Figure 3 shows the initial re- sults from the scene understanding module where the two users are tracked in real-time and their names … For natural language under- standing, we use Semantic Interpretation for Speech Recognition (SISR)3 in link with the SRGS grammars. … Related articles

The SUN Action database: collecting and analyzing typical actions for visual scene types CAW Olsson – 2013 – dspace.mit.edu … I introduce the SUN Action database (short for “Scene UNderstanding – Action”): the first effort to collect and analyze free-response data from human subjects about the typical ac- tions associated with different scene types … 6.6 Incorporate semantic properties of natural language . … Related articles All 2 versions

Research on the Text Detection and Extraction from Complex Images J Zhang, R Cheng, K Wang… – Emerging Intelligent Data …, 2013 – ieeexplore.ieee.org … Initially, this model was applied to natural language processing. … Learning a classification model for segmentation.ICCV 2003:10 – 17 vol.1 [57] Kumar, MP Efficiently selecting regions for scene understanding CVPR 2010: 3217 – 3224 [58] Michael Mairel, Stella X. Yu2, Pietro … Cited by 1 Related articles All 4 versions

Semantic classification of scenes and places with omnidirectional vision LF Posada, KK Narayanan, F Hoffmann… – … Robots (ECMR), 2013 …, 2013 – ieeexplore.ieee.org … semantic understanding of the environments in which they operate in order to comprehend human task descriptions in natural language(eg go … recognizing the overall scene, they lack a segmentation of the scene into objects required for a more profound scene understanding. … Cited by 1 Related articles

2013 Workshop on Computational Models of Narrative MA Finlayson, B Fisseni, B Löwe, JC Meister – 2013 – vesta.informatik.rwth-aachen.de … General, I.2.1 Applications and Expert Systems, I.2.4 Knowledge Representation Formalisms and Methods, I.2.10 Vision and Scene Understanding, I.2.11 Distributed Artificial Intelligence, J.5 Literature, I.2 Artificial Intelligence, I.2.6 Learning, I.2.7 Natural Language Processing, I … All 2 versions

SpeedMachines: Anytime Structured Prediction A Grubb, D Munoz, JA Bagnell, M Hebert – arXiv preprint arXiv:1312.0579, 2013 – arxiv.org … 4 Anytime Scene Understanding 4.1 Background In addition to part-of-speech tagging in natural language processing, scene un- derstanding in computer vision is another important and challenging structured prediction problem. … Related articles All 5 versions

A General Framework for Recognizing Complex Events in Markov Logic. YC Song, HA Kautz, Y Li, J Luo – AAAI Workshop: Statistical Relational …, 2013 – aaai.org … There is growing interest in machine learn- ing and computational linguistics in models that unify visual perception and natural language processing. … scene understanding using probabilistic logic models and common sense reasoning. … Related articles All 15 versions

Defining of printing plate quality level through introduction of different image analysis applications J Poljak – 2013 – eprints.grf.hr … In that way the artificial intelligence is developed and systems acquire the necessary skills such as: ability to communicate in natural language, the ability to learn and … object-based (model-based) compression of video streams • general scene understanding [17]. Page 19. 13 … Cited by 1 Related articles All 2 versions

User Interface Patterns for Multimodal Interaction A Ratzka – Transactions on Pattern Languages of Programming III, 2013 – Springer … PhD thesis, School of Computer Science, Computer Science Department, Carnegie Mellon University (1998); Cohen, PR: The role of natural language in a multimodal … ACM Press, New York (2004); Wachsmuth, S.: Multi-modal Scene Understanding Using Probabilistic Models. … Cited by 1 Related articles All 4 versions

Towards Cognitive Image Interpretation by Mixing Qualitative Image Descriptions and Feature Object Detectors Z Falomir, L Museros, L Gonzalez-Abril… – 27th International …, 2013 – upcommons.upc.edu … tion and scene understanding from a 3D perspective is also intended by interpreting the point clouds given as the output of RGB-Depth … intend to:(i) create a repository of tar- get objects for benchmarking;(ii) define a grammar for gen- erating natural language descriptions from … Related articles

Exploring the effects of gaze and pauses in situated human-robot interaction G Skantze, A Hjalmarsson, C Oertel – … of the 14th Annual Meeting of …, 2013 – sigdial.org Page 1. Proceedings of the SIGDIAL 2013 Conference, pages 163–172, Metz, France, 22-24 August 2013. c 2013 Association for Computational Linguistics Exploring the effects of gaze and pauses in situated human-robot interaction … Cited by 8 Related articles All 8 versions

Object Recognition System Design in Computer Vision: a Universal Approach A Gleibman – arXiv preprint arXiv:1310.7170, 2013 – arxiv.org … This kind of logic provides a way to build natural language processing applications immediately from examples without the concept of formal … In order to build a specific application, the univer- sal system can be trained for performing specific scene understanding tasks and then … Cited by 1 Related articles All 2 versions

Performance evaluation of Different Annotation retrieval methods N Sharma, G Jindal – ijiet.com … The description of medical data is normally presented in terms of controlled, but diversely expressedvocabulary and natural language specifications. … General object recognition and scene understanding techniques are used to extract the semantics from data. … Related articles

Comparative Analysis of Semantic Search and Annotation retrieval N Sharma, G Jindal – ijiet.com … The description of medical data is normally presented in terms of controlled, but diversely expressed vocabulary and natural language specifications. … General object recognition and scene understanding techniques are used to extract the semantics from data. … Related articles All 3 versions

Adaptive Visual Tracking via Learning Detector of Specific Landmarks CL Hwang, KC Chang – Computational Intelligence and Virtual …, 2013 – ieeexplore.ieee.org … This is a reasonable compromise between the simplicity of the parameterization and the subsequent scene understanding. … 1998. [16] D. Pierce and C. Cardie, “Limitations of co-training for natural language learning from large datasets,” The Conference on Empirical Methods in … Related articles

Shopping behavior recognition using a language modeling analogy MC Popa, LJM Rothkrantz, P Wiggers… – Pattern Recognition Letters, 2013 – Elsevier … In this work, we address the analogy between human shopping behavior and a natural language. … Visual scene understanding and human behavior recognition is also affected by contextual knowledge and could benefit a lot from this type of information. … Cited by 2 Related articles All 5 versions

Intelligent interactive multimedia system: layered ontological video contexts in a folksonomy driven environment M Dal Mas – Proceedings of the 3rd International Conference on …, 2013 – dl.acm.org … I.2.2, I.2.3, I.2.4); H.1.2 User/Machine Systems; H.5.1 Multimedia Information Systems; H.5.2 Graphical user interfaces (GUI); I.2.4 Knowledge Representation (F.4.1); I.2.6 Knowledge acquisition; I.2.7 Natural Language Processing; I.2.10 Vision and Scene Understanding; I.4.8 … Related articles All 6 versions

Recurrent Convolutional Neural Networks for Scene Parsing PHO Pinheiro, R Collobert – arXiv preprint arXiv:1306.2795, 2013 – arxiv.org … IEEE 12th International Conference on Computer Vision (ICCV) (2009) 2. Kumar, MP and Koller, D.: Efficiently selecting regions for scene understanding. … 2009) 6. Socher, R. and Lin, C. and Ng, A. and Manning, C.: Parsing Natural Scenes and Natural Language with Recursive … Cited by 2 Related articles All 4 versions

Indoor scene recognition by a mobile robot through adaptive object detection P Espinace, T Kollar, N Roy, A Soto – Robotics and Autonomous Systems, 2013 – Elsevier … 10] and [11]. In terms of semantic understanding of an environment, scene recognition appears as one of the most fundamental perceptual abilities, being a key contextual cue for scene understanding. Consequently, there is … Cited by 6 Related articles All 4 versions

[BOOK] Automatic Camera Selection for Format Agnostic Live Event Broadcast Production R Kaiser, W Weiss, G Kienast, W Bailer, M Thaler… – joanneum.at … In order to define the PSE’s intended behaviour; we observe TV broadcasts and interview professionals. The production rules are initially captured in natural language before they are developed as JBoss Drools rules. … 4 Scene Understanding … Related articles

Deep learning of representations: Looking forward Y Bengio – Statistical Language and Speech Processing, 2013 – Springer … However, whereas the task of recognizing handwritten digits is solved to the point of achieving roughly human-level performance, this is far from true for tasks such as general object recognition, scene understanding, speech recognition, or natural language understanding. … Cited by 43 Related articles All 7 versions

A Hierarchical Segmentation Approach towards Roads and Slopes for Collapse Recognition. Y Jia, Z He, H Zhu – International Journal of Signal Processing, Image …, 2013 – sersc.org … Symposium, Eindhoven, Netherlands, (2008) June 4-6. [5] A. Tsukada, M. Ogawa and F. Galpin, “Road structure based scene understanding for intelligent … His research interests are machine learning, pattern recognition, image processing and natural language processing. … Related articles All 3 versions

Hierarchical crowd analysis and anomaly detection X Chong, W Liu, P Huang, NI Badler – Journal of Visual Languages & …, 2013 – Elsevier … Wang et al. [14] and [15] introduced natural language processing (NLP) into dealing with trajectory analysis, which clustered trajectories by using nonparametric Bayesian model to find atomic activities and build semantic regions. Saligrama et al. … Cited by 1 Related articles

Route description interpretation on automatically labeled robot maps C Landsiedel, R de Nijs, K Kuhnlenz… – … (ICRA), 2013 IEEE …, 2013 – ieeexplore.ieee.org … [2] T. Kollar, S. Tellex, and N. Roy, “A discriminative model for under- standing natural language route directions,” in … [5] M. Johnson-Roberson, J. Bohg, G. Skantze, J. Gustafson, R. Carlson, B. Rasolzadeh, and D. Kragic, “Enhanced visual scene understanding through human … Related articles All 2 versions

Image categorization using a semantic hierarchy model with sparse set of salient regions C Liu, Y Zheng, S Gong – Frontiers of Computer Science, 2013 – Springer … Assuming that each image is generated from multiple la- tent topics, latent dirichlet allocation (LDA) from natural language processing has been successfully applied to image and video content analysis using the bag of words model, in which “visual words” are taken as atomic … Related articles All 3 versions

Efficient temporal consistency for streaming video scene analysis O Miksik, D Munoz, JA Bagnell… – … and Automation (ICRA), …, 2013 – ieeexplore.ieee.org … margin markov networks,” in CVPR, 2009. [3] R. Socher, CC-Y. Lin, AY Ng, and CD Manning, “Parsing natural scenes and natural language with recursive neural networks,” in ICML, 2011. [4] C. Farabet, C. Couprie, L. Najman … Cited by 12 Related articles All 13 versions

Beyond myopic inference in big data pipelines K Raman, A Swaminathan, J Gehrke… – Proceedings of the 19th …, 2013 – dl.acm.org … creation, model construction, testing, and visualization. In order not to have to develop components for each task from scratch, data scientists build Big Data Pipelines using exist- ing software components, for example, a natural language parser or a text classifier. … Related articles All 6 versions

Semantic, Automatic Image Annotation Based On Multi-Layered Active Contours and Decision Trees JI Olszewska – … Journal of Advanced Computer Science and …, 2013 – eprints.hud.ac.uk … Index Terms—Automatic Image Annotation, Natural Language Tags, Decision Trees, Semantic Attributes, Visual Features, Active Contours, Segmentation, Image … concepts have been proven to complement well visual information in the process of scene understanding [23]. … Related articles All 8 versions

Visual Search: A Large-Scale Perspective R Piramuthu, A Bhardwaj, W Di… – Handbook of Statistics: …, 2013 – books.google.com … Most common cues include color, texture, shape, depth (vanishing points, occlusion), illumination and shading, grouping (similar color, texture, proximity, common fate), parts/attributes, 3D geometry, geometric layout, scene understanding, activity understanding, tracking, and … Related articles

Optimizing Nondecomposable Loss Functions in Structured Prediction M Ranjbar, T Lan, Y Wang… – Pattern Analysis and …, 2013 – ieeexplore.ieee.org … Examples include performance measures such as F score (natural language processing), intersection over union (object category … Tasks such as machine translation, object category segmentation, and scene understanding involve reasoning about relationships between words … Related articles All 14 versions

Unsupervised hierarchical structure induction for deeper semantic analysis of audio S Chaudhuri, B Raj – Acoustics, Speech and Signal Processing …, 2013 – ieeexplore.ieee.org … Besides, while the presented approach makes no claims toward modeling human approaches to scene understanding, the extracted co-occurrence information and … to the similarity of the task we are attempting to solve with those from the parsing of natural language text, the … Cited by 5 Related articles

The ethics of artificial intelligence N Bostrom, E Yudkowsky – Cambridge Handbook of Artificial …, 2013 – books.google.com Page 329. 15 15.1 The ethics of artificial intelligence Nick Bostrom and Eliezer Yudkowsky The possibility of creating thinking machines raises a host of ethical issues, related both to ensuring that such machines do not harm … Cited by 18 Related articles All 12 versions

Indoor Place Classification for Intelligent Mobile Systems L Shi – 2013 – opus.lib.uts.edu.au … concepts connects the research on high-level cognition (ie human-robot interaction) like natural language processing (NLP), and the research on intelligent mobile systems (ie … Scene analysis, also known as scene understanding, image parsing or image …

Superparsing J Tighe, S Lazebnik – International Journal of Computer Vision, 2013 – Springer … problem. Finally, we present an exten- sion of our method to video sequences and report results on a video dataset with frames densely labeled at 1 Hz. Keywords Scene understanding · Image parsing · Image segmentation J … Cited by 44 Related articles All 9 versions

Abstraction in Different Disciplines L Saitta, JD Zucker – Abstraction in Artificial Intelligence and Complex …, 2013 – Springer … 2.2 Natural Language. As we have already mentioned, in natural languages there is a distinction between abstract and concrete words. … A signal system, instead, can only communicate what is here and now. In natural language abstraction enters also as a figure of style. …

Function-Theoretic Explanation and Neural Mechanisms F Egan – philpapers.org … seeing 3D structure in the scene, understanding speech, grasping objects in nearby space, and … 17 The computational model of natural language processing presented by Marcus (1980) is a case in point: the proposed model is an argumentation of a standard LR(k) parser of … Related articles All 3 versions

Spatial inference machines R Shapovalov, D Vetrov, P Kohli – Computer Vision and Pattern …, 2013 – ieeexplore.ieee.org … They are used to solve a wide range of labelling problems encoun- tered in computer vision (semantic segmentation [5, 7, 11], dense stereo estimation [18], denoising [20, 13]) and be- yond (part of speech tagging in natural language process- ing, gene finding in bioinformatics … Cited by 7 Related articles All 13 versions

Comprehending Spatial Relations for Interactive Object Recognition L Cao – 2013 – sucra.saitama-u.ac.jp … We propose a novel model to encode linguistic spatial expressions. We first investigated how humans manipulate space by the action of natural language and classified basic … 81 5.2 The role of Natural Language ….. 81 …

The Scope and Limits of Simulation in Cognition and Automated Reasoning E Davis, G Marcus – Under submission, 2013 – cs.nyu.edu … It has often been argued that simulation plays a central role in a wide range of cognitive tasks, including physical reasoning and natural language understanding, and correspondingly that simulation should be used as the basis for carrying out comparable tasks in artificial … Cited by 1 Related articles

New approach for human detection in images using histograms of oriented gradients NG Bardeh, M Palhang – Electrical Engineering (ICEE), 2013 …, 2013 – ieeexplore.ieee.org … B. Bag of Feature In the proposed method, we use some concepts of natural language processing and data mining. … 90-97. [10] L. Ladicky, “Global Structured Models towards Scene Understanding”, Ph.D Dissertaion, Dept. Engineering Sciences, Univ. Oxford Brookes, 2011. … Cited by 1 Related articles All 2 versions

Adaptive context exploitation AN Steinberg, CL Bowman – SPIE Defense, …, 2013 – proceedings.spiedigitallibrary.org … An adaptive evidence-accrual inference method – adapted from developments in target recognition and scene understanding – is presented … that establishes the pertinent context for predicting and understanding (ie characterizing the meaning of) natural language utterances. … Cited by 3 Related articles All 3 versions

Task-specific image partitioning S Kim, S Nowozin, P Kohli… – Image Processing, IEEE …, 2013 – ieeexplore.ieee.org … I. INTRODUCTION REGION-BASED image representations (RBIRs) have been shown to be effective in improving the performance of algorithms for high-level image/scene understanding, which encompasses tasks such as object class segmentation, scene segmentation … Cited by 12 Related articles All 15 versions

Attribute-based classification for zero-shot learning of object categories C Lampert, H Nickisch, S Harmeling – 2013 – ieeexplore.ieee.org … Index Terms—Object recognition, vision and scene understanding Ç … [50] used attribute predictions in combination with object detection and techniques from natural language processing to automatically create descriptions of images in natural language of images. … Cited by 24 Related articles All 11 versions

Segmentation based classification of 3D urban point clouds: A super-voxel based approach with evaluation AK Aijazi, P Checchin, L Trassoudaine – Remote Sensing, 2013 – mdpi.com … 2.1. Segmentation of 3D Data. In order to fully exploit 3D point clouds, for scene understanding and object classification, effective segmentation has proved to be a necessary and critical pre-processing step in a number of autonomous perception tasks. … Cited by 7 Related articles All 5 versions

[BOOK] Language, Music, and the Brain: A Mysterious Relationship MA Arbib – 2013 – books.google.com Page 1. Language, Music, and the Brain A Mysterious Relationship EDITED BY Michael A. Arbib STRUNGMANN FORUM REPORTS Page 2. Language, Music, and the Brain A Mysterious Relationship Page 3. Striingmann Forum … Cited by 13 Related articles

Machine Vision: Metrics for Evaluating VMIATO EVALUATE – Echnology Assessment in Software …, 2013 – books.google.com … and very much ad hoc. All of the high-level processes are derived from established artificial intelligence concepts (see Ballard & Brown, 1982) that have been borrowed from natural language processing research. It is not clear … Related articles

Supervised language models for temporal resolution of text in absence of explicit temporal cues A Kumar – 2013 – repositories.lib.utexas.edu … 1.3 The Approach Our primary focus here is the challenge of learning implicit temporal distributions over natural language use at-large. As a concrete example, consider the word wireless. Introduced in the early 20th century to refer to radios, it fell into disuse and then its … Related articles

Modeling and visual recognition of human actions and interactions I Laptev – 2013 – tel.archives-ouvertes.fr … definition. The challenge of defining an action vocabulary has also been recognized in natural language processing where recent studies explore eg, the problem of grounding natural language to robot control commands [53]. …

On sampling from the gibbs distribution with random maximum a-posteriori perturbations T Hazan, S Maji, T Jaakkola – Advances in Neural Information …, 2013 – papers.nips.cc … 1 Introduction Inference in complex models drives much of the research in machine learning applications, from computer vision, natural language processing, to computational biology. Examples include scene understanding, parsing, or protein design. … Cited by 8 Related articles All 8 versions

Virtual Director R Kaiser, W Weiss – Media Production, Delivery and Interaction …, 2013 – books.google.com … advanced feature, yet quickly required in professional productions, is the ability to go beyond a posteriori scene understanding by pre … Human broadcast professionals are usually able to express desired behaviour in natural language rules resembling ECA format, which can be … Related articles All 2 versions

Attribit: content creation with semantic attributes S Chaudhuri, E Kalogerakis, S Giguere… – Proceedings of the 26th …, 2013 – dl.acm.org … ACM Classification Keywords H.5.2 Information Interfaces and Presentation (eg, HCI): User Interfaces – Interaction styles; I.2.10 Artificial Intelli- gence: Vision and Scene Understanding – Shape General Terms Algorithms; Design; Human Factors. … Cited by 5 Related articles All 8 versions

Learning convex QP relaxations for structured prediction J Jancsary, S Nowozin, C Rother – … of The 30th International Conference on …, 2013 – jmlr.org Page 1. Learning Convex QP Relaxations for Structured Prediction Jeremy Jancsary jermyj@microsoft.com Sebastian Nowozin senowozi@microsoft.com Carsten Rother carrot@microsoft.com Microsoft Research Cambridge … Cited by 1 Related articles All 12 versions

Productive Vision: Methods for Automatic Image Comprehension DA Summers-Stay – 2013 – drum.lib.umd.edu … possible labelings. • Answer questions about the scene that require inference and give justifications for the answers in natural language. • Create a visual representation of the scene in a new medium. … 11 3 Visual Filters for Scene Understanding 14 …

Interactive Learning for Sequential Decisions and Predictions S Ross – 2013 – gradworks.umi.com … are exhibited over a large variety of challenging learning tasks, ranging from learning video game playing agents from human players and accurate dynamic mod els of a simulated helicopter for controller synthesis, to learning predictors for scene understanding in computer … Cited by 2 Related articles All 2 versions

Sudden event recognition: A survey NS Suriani, A Hussain, MA Zulkifley – Sensors, 2013 – mdpi.com … and predefined normal behaviors. Jeong and Yang [79] presented each target activity as ground, weighted and undirected trees. Snatching activity is defined in natural language and the Horn clause. For example, a snatch event … Cited by 2 Related articles All 12 versions

Knowledge Based Perceptual Anchoring: Grounding percepts to concepts in cognitive robots M Daoutis – Künstliche Intelligenz, 2013 – aass.oru.se Page 1. Doctoral Dissertation Knowledge Based Perceptual Anchoring: Grounding percepts to concepts in cognitive robots Marios Daoutis Computer Science Örebro Studies in Technology 55 Örebro 2013 Page 2. Page 3. Knowledge Based Perceptual Anchoring: Grounding … Cited by 5 Related articles All 7 versions

Defining and evaluating qualitative spatialrelations for describing desktop scenes A Gallart del Burgo – 2013 – diva-portal.org Page 1. ¿ Degree project in Computer Science Second cycle Stockholm, Sweden 2013 Defining and evaluating qualitative spatial relations for describing desktop scenes Adrià Gallart Del Burgo Page 2. Defining and evaluating qualitative spatial … Related articles All 2 versions

Labeling Spain With Stanford Y Zhou, I Nwogu, V Govindaraju – 2013 – ieeexplore.ieee.org … (images best viewed in color) A. Related work In many high-level scene understanding tasks, it is important to specifically extract the regions present in a scene, in order to reason more accurately about the high-level concepts that exist in that scene. … Related articles All 7 versions

Generalized World Entities as an Unifying IoT Framework: A Case for the GENIUS Project GP Zarri – Internet of Things and Inter-cooperative Computational …, 2013 – Springer … high-level semantic knowledge in different functions of robots such as: mapping and localization, scene understanding, smart and … These links are normally expressed in natural language through syntactic constructions like causality, goal, indirect speech, co-ordination and … Related articles All 2 versions

Learning multi-modal latent attributes Y Fu, T Hospedales, T Xiang, S Gong – 2013 – ieeexplore.ieee.org Page 1. Learning Multimodal Latent Attributes Yanwei Fu, Student Member, IEEE, Timothy M. Hospedales, Member, IEEE, Tao Xiang, and Shaogang Gong Abstract—The rapid development of social media sharing has created … Cited by 14 Related articles All 12 versions

Beyond physical product configuration–Configuration in unusual domains L Hotz, K Wolter – AI Communications, 2013 – IOS Press Page 1. AI Communications 26 (2013) 39–66 39 DOI 10.3233/AIC-2012-0545 IOS Press Beyond physical product configuration – Configuration in unusual domains Lothar Hotz * and Katharina Wolter HITeC eV c/o Fachbereich … Cited by 2 Related articles All 3 versions

Graphical Model Inference and Learning for Visual Computing N Komodakis – 2013 – tel.archives-ouvertes.fr … Note, for instance, that advances in scene understanding can in turn bring significant improvements in areas as diverse as autonomous robots … applicable to a number of other areas of artificial intelligence as well, such as, for instance, natural language processing, computational … Related articles All 9 versions

Teleoperation and beyond for assistive humanoid robots MA Goodrich, JW Crandall… – Reviews of Human …, 2013 – rev.sagepub.com … Wong, Xu, & Saxena, 2011; Kennedy et al., 2007), (c) prediction and plan- ning, and (d) natural language understanding (Brick … 2009), gesture recognition (Corradini, 2000; Stiefelhagen et al., 2004), object recognition (Roth & Winter, 2008), and scene understanding (Li, Socher … Cited by 2 Related articles All 2 versions

[BOOK] Abstraction in artificial intelligence and complex systems L Saitta, JD Zucker – 2013 – Springer Page 1. Lorenza Saitta Jean-Daniel Zucker Abstraction in Artificial Intelligence and Complex Systems Page 2. Abstraction in Artificial Intelligence and Complex Systems Page 3. (Pierre Soulages, Peinture 260 9 202 cm, 19 juin 1963) … Cited by 3 Related articles All 6 versions

2013 Index IEEE Transactions on Pattern Analysis and Machine Intelligence Vol. 35 C Alberola-Lopez, O Aldrian, A Alonso-Betanzos… – ieeexplore.ieee.org Page 1. IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 36, NO. 1, JANUARY 2104 1 2013 Index IEEE Transactions on Pattern Analysis and Machine Intelligence Vol. 35 This index covers … All 2 versions

[BOOK] Qualitative Constraint Satisfaction Problems: Algorithms, Computational Complexity, and Extended Framework W Liu – 2013 – epress.lib.uts.edu.au … intelligent agents, and in disciplines including robot planning and navigation, natu- ral language understanding, scene understanding and even fields outside AI (such as … high-level computer vision, natural language processing, cognitive systems [79], doc- … Related articles

Intuitive Human-Robot Interaction by Intention Recognition M Awais – 2013 – epub.uni-bayreuth.de … specific HRI scenario. The presented approach does not propose an image- processing-based method for scene understanding. The process of scene understanding corresponds to the apprehension of scene. The approach … Related articles All 5 versions

Knowledge-based 3D point clouds processing QH Truong – 2013 – hal.archives-ouvertes.fr Page 1. Thèse de Doctorat ¦ école doctorale sciences pour l’ingénieur et microtechniques UNIVERSIT É DEBOURGOGNE Knowledge-based 3D point clouds processing Traitement 3D de nuages de points bas´e sur la connaissance Quoc Hung TRUONG … Related articles All 7 versions

Election Data Visualisation E Long – 2013 – pearl.plymouth.ac.uk … 205 Figure 8.7: Example of result of the 2010 general election ….. 206 Figure 8.8: Screenshot of the Natural Language Enquiry Template ….. 208 Figure 8.9: Screenshot of the Interface for Enquiry Creation ….. … Cited by 1 Related articles All 9 versions

2012 Index IEEE Transactions on Pattern Analysis and Machine Intelligence Vol. 34 T Aach, R Achanta, A Adam, L Agapito… – … on Pattern Analysis …, 2013 – ieeexplore.ieee.org … 2012 2489-2495 Li, Congcong, Kowdle, Adarsh, Saxena, Ashutosh, and Chen, Tsuhan, Toward Holistic Scene Understanding: Feedback Enabled Cascaded Classification Models; TPAMI July 2012 1394-1408 Li, Hongdong, see Hartley, Richard, TPAMI Dec. … All 4 versions

Latent Structured Active Learning W Luo, A Schwing, R Urtasun – Advances in Neural Information …, 2013 – papers.nips.cc … For example, in natural language processing, we might be interested in parsing sentences syntactically. In computer vision, we might want to predict the depth of each pixel, or its semantic category. … Efficient Structured Prediction for 3D Indoor Scene Understanding. In Proc. … Cited by 1 Related articles All 10 versions

Classification and Counting of Composite Objects in Traffic Scenes Using Global and Local Image Analysis G Somasundaram, R Sivalingam… – Intelligent …, 2013 – ieeexplore.ieee.org … A. BoVW Classifier The bag-of-visual-words (BoVW) method for modeling ob- jects is derived from the general principles of natural language processing. The model learns “words” or meaningful features of different object classes. The general distribution of these … Cited by 1 Related articles All 3 versions

[BOOK] Informatics in Radiation Oncology G Starkschall, RAC Siochi – 2013 – books.google.com … 2007). Software development can include knowl- edge areas, such as natural language processing, neural networks, clustering analysis, genetic algorithms, Monte Carlo simula- tions, and many others (Altman 1998). Along … Related articles All 3 versions

Batch Mode Active Learning for Multimedia Pattern Recognition S Chakraborty – 2013 – repository.asu.edu Page 1. Batch Mode Active Learning for Multimedia Pattern Recognition by Shayok Chakraborty A Dissertation Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy Approved April 2013 by the Graduate Supervisory Committee: … Related articles All 6 versions

Detecting and indexing moving objects for Behavior Analysis by Video and Audio Interpretation LUC BRUN – 2013 – elcvia.cvc.uab.es Page 1. Università degli Studi di Salerno Dipartimento di Ingegneria dell’Informazione, Ingegneria Elettrica e Matematica Applicata Dottorato di Ricerca in Ingegneria dell’Informazione XII Ciclo – NS Doctorat de l’université de Caen Basse-Normandie PHD THESIS …

Continuous-time infinite dynamic topic models W Elshamy – arXiv preprint arXiv:1302.7088, 2013 – arxiv.org Page 1. CONTINUOUS-TIME INFINITE DYNAMIC TOPIC MODELS by WESAM SAMY ELSHAMY BS, Ain Shams University, Egypt, 2004 M.Sc., Cairo University, Egypt, 2007 AN ABSTRACT OF A DISSERTATION submitted in partial fulfillment of the requirements for the degree … Related articles All 9 versions

Learning Everything about Anything: Webly-Supervised Visual Concept Learning SK Divvala, A Farhadi, C Guestrin – ai-squared.org Page 1. Learning Everything about Anything: Webly-Supervised Visual Concept Learning Santosh K. Divvala*,†, Ali Farhadi†, Carlos Guestrin† †University of Washington *The Allen Institute for AI http://levan.cs.uw.edu Figure … Cited by 1 Related articles All 3 versions

GraphLab: A Distributed Abstraction for Large Scale Machine Learning Y Low – 2013 – cs.cmu.edu Page 1. GraphLab: A Distributed Abstraction for Large Scale Machine Learning Yucheng Low June 2013 Machine Learning Department School of Computer Science Carnegie Mellon University Pittsburgh, PA 15213 Thesis … Cited by 3 Related articles All 7 versions

[BOOK] Vision-based Pedestrian Protection Systems for Intelligent Vehicles D Gerónimo, AM López – 2013 – Springer Page 1. SPRINGER BRIEFS IN COMPUTER SCIENCE David Gerónimo · Antonio M. López Vision-based Pedestrian Protection Systems for Intelligent Vehicles Page 2. SpringerBriefs in Computer Science Series Editors Stan … Cited by 2 Related articles All 4 versions

Bayesian nonparametric approach to blind separation of infinitely many sparse sources H Kameoka, S SAGAYAMA – IEICE Transactions on Fundamentals …, 2013 – search.ieice.org Page 1. 1928 IEICE TRANS. FUNDAMENTALS, VOL.E96–A, NO.10 OCTOBER 2013 PAPER Special Section on Sparsity-aware Signal Processing Bayesian Nonparametric Approach to Blind Separation of Infinitely Many Sparse Sources … Cited by 3 Related articles All 4 versions

Associative Hierarchical Random Fields L Ladicky, C Russell, P Kohli, P Torr – 2013 – ieeexplore.ieee.org Page 1. 1056 IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE, VOL. 36, NO. 6, JUNE 2014 Associative Hierarchical Random Fields L’ubor Ladický, Chris Russell, Pushmeet Kohli, and Philip HS Torr … Cited by 7 Related articles All 6 versions

Domain Adaptation of Deformable Part-Based Models J Xu, S Ramos, D Vázquez, AM López – cvc.uab.es Page 1. 1 Domain Adaptation of Deformable Part-Based Models Jiaolong Xu, Student Member, IEEE, Sebastian Ramos, Student Member, IEEE, David Vázquez, Member, IEEE, Antonio M. López, Member, IEEE Abstract—The … Cited by 2 Related articles All 3 versions

Virtual and real world adaptation for pedestrian detection D Vazquez, A López, J Marin, D Ponsa, D Gerónimo – 2013 – ieeexplore.ieee.org … This problem is referred to as data set shift [27] and is receiving increasing atten- tion in the Machine Learning field [27], [28], [29] due to its relevance in areas like natural language processing, speech processing, and brain-computer interfaces, to mention a few. … Cited by 10 Related articles All 6 versions