Filters








23,474 Hits in 3.9 sec

Anchoring in a grounded layered architecture with integrated reasoning

Stuart C. Shapiro, Haythem O. Ismail
2003 Robotics and Autonomous Systems  
The GLAIR grounded layered architecture with integrated reasoning for cognitive robots and intelligent autonomous agents has been used in a series of projects in which Cassie, the SNePS cognitive agent  ...  We discuss anchoring in the domains of: perceivable entities and properties, actions, time, and language.  ...  We also appreciate the efforts and discussions of Silvia Coradeschi, Alessandro Saffiotti, and other organizers and attendees of the 2001 AAAI Fall Symposium on Anchoring Symbols to Sensor Data in Single  ... 
doi:10.1016/s0921-8890(02)00352-4 fatcat:mfaobnj55nhaxngauzvjxbynp4

Grounding commonsense knowledge in intelligent systems

Marios Daoutis, Silvia Coradeshi, Amy Loutfi
2009 Journal of Ambient Intelligence and Smart Environments  
This first integration work is a step towards integrating the richness of a KRR system developed over many years in isolation, with a physically embedded intelligent system.  ...  In this work we present an integration between ResearchCyc and an anchoring framework that mediates the connection between the perceptual information in an intelligent home environment and the reasoning  ...  Namely there is a perceptual layer, a symbol grounding and maintenance layer, and finally the Knowledge Representation and Reasoning Layer.  ... 
doi:10.3233/ais-2009-0040 fatcat:qxpkgp4ytvel7oe6mgzynckm4u

Box-level Segmentation Supervised Deep Neural Networks for Accurate and Real-time Multispectral Pedestrian Detection [article]

Yanpeng Cao, Dayan Guan, Yulun Wu, Jiangxin Yang, Yanlong Cao, Michael Ying Yang
2019 arXiv   pre-print
In this paper, we present a novel box-level segmentation supervised learning framework for accurate and real-time multispectral pedestrian detection by incorporating features extracted in visible and infrared  ...  It offers two major advantages over the existing anchor box based multispectral detection methods.  ...  Network Architecture Box-level segmentation for Supervised Training A common step of state-of-the-art anchor box based detectors is generating a large number of anchor boxes with various sizes and aspect  ... 
arXiv:1902.05291v1 fatcat:h7qdwr6g3rfm7iqjyaaf5dduvy

Towards integrated neural–symbolic systems for human-level AI: Two research programs helping to bridge the gaps

Tarek R. Besold, Kai-Uwe Kühnberger
2015 Biologically Inspired Cognitive Architectures  
The second program suggests a new approach and computational architecture for the cognitively-inspired anchoring of an agent's learning, knowledge formation, and higher reasoning abilities in real-world  ...  Abstract After a Human-Level AI-oriented overview of the status quo in neural-symbolic integration, two research programs aiming at overcoming long-standing challenges in the field are suggested to the  ...  With respect to the research program on complexity differences between paradigms, a debt of gratitude is owed to Frank Jäkel (University of Osnabrück) 880 for repeated rounds of feedback and many valuable  ... 
doi:10.1016/j.bica.2015.09.003 fatcat:bwp3ljsconfrzj4lfaazjhbf6u

3D Object Detection on Point Clouds using Local Ground-aware and Adaptive Representation of scenes' surface [article]

Arun CS Kumar, Disha Ahuja, Ashwath Aithal
2020 arXiv   pre-print
A novel, adaptive ground-aware, and cost-effective 3D Object Detection pipeline is proposed.  ...  The ground surface representation introduced in this paper, in comparison to its uni-planar counterparts (methods that model the surface of a whole 3D scene using single plane), is far more accurate while  ...  , in a two-staged architecture.  ... 
arXiv:2002.00336v2 fatcat:6vvpi4jg4rbrfcixinp5s7lpxe

LCR-Net: Localization-Classification-Regression for Human Pose

Gregory Rogez, Philippe Weinzaepfel, Cordelia Schmid
2017 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
We propose an end-to-end architecture for joint 2D and 3D human pose estimation in natural images.  ...  Our architecture, named LCR-Net, contains 3 main components: 1) the pose proposal generator that suggests potential poses at different locations in the image; 2) a classifier that scores the different  ...  If B has a high overlap with several poses, let (p, P ) be the ground-truth pose with the highest IoU with the box.  ... 
doi:10.1109/cvpr.2017.134 dblp:conf/cvpr/RogezWS17 fatcat:lwvkswnrbzgvnmri2nbsjpihty

Utilizing RFID-WSNs for reducing the footprint of the Oil Sands industry

Ashraf Al-Fagih, Ahmad El Kouche, Sharief Oteafy, Abdallah Alma'aitah
2012 2012 IEEE Globecom Workshops  
This paper proposes the use of an integrated RFID-WSN architecture to reduce the footprint of mining equipment used in the Canadian Oil Sands industry.  ...  In harsh operational conditions, machines lose components on a frequent basis, causing a significant footprint to the surrounding environment.  ...  Network Model We adopt a four-layered hierarchical architecture for detecting broken Ground Engaging Tools (GETs) in the Oil Sands mining locations of northern Canada.  ... 
doi:10.1109/glocomw.2012.6477600 dblp:conf/globecom/Al-FagihKOA12 fatcat:oaihs3vy6jhijduik5anoedmpa

Integrated Deep Model for Face Detection and Landmark Localization From "In The Wild" Images

Gary Storey, Ahmed Bouridane, Richard Jiang
2018 IEEE Access  
MODEL TRAINING Our IDM method specifically trains the layers of a faster R-CNN architecture [4] with images containing multiple faces from the Wider Face training set [57] .  ...  METHOD The proposed Integrated Deep Model for both face detection and landmark localisation from 'in the wild' images integrates two state-of-the-art architectures namely the Faster R-CNN network architecture  ... 
doi:10.1109/access.2018.2882227 fatcat:t4t74wtktbdurhra7igoj7rrle

The potential of cooperative communications to speed up disaster relief operations

Farouk Mezghani, Nathalie Mitton
2019 2019 International Conference on Information and Communication Technologies for Disaster Management (ICT-DM)  
The documents may come from teaching and research institutions in France or abroad, or from public or private research centers.  ...  HAL is a multi-disciplinary open access archive for the deposit and dissemination of scientific research documents, whether they are published or not.  ...  This is thus worth integrating in our scheme the fact that all devices can represent the clique for a given layer. B.  ... 
doi:10.1109/ict-dm47966.2019.9032963 dblp:conf/ict-dm/MezghaniM19 fatcat:cmio5yy5yvg3dp36hnm7okwub4

Attention Fusion for One-Stage Multispectral Pedestrian Detection

Zhiwei Cao, Huihua Yang, Juan Zhao, Shuhong Guo, Lingqiao Li
2021 Sensors  
The log-average miss rate (MR) for the two modalities with reasonable settings were 4.91% and 23.14%, respectively.  ...  in the two modalities.  ...  However, in the fully connected layer, each node in the output layer connects directly to a node in the previous layer.  ... 
doi:10.3390/s21124184 fatcat:uiw3biqjofcifgt7fwp3o22gr4

Semantic Relational Object Tracking [article]

Andreas Persson, Pedro Zuidberg Dos Martires, Amy Loutfi, Luc De Raedt
2019 arXiv   pre-print
For more complex scenarios, a high-level probabilistic object tracker has been integrated with the anchoring framework and handles the tracking of occluded objects via reasoning about the state of unobserved  ...  A novel anchoring matching function method learns to maintain object entities in space and time and is validated using a large set of trained humanly annotated ground truth data of real-world objects.  ...  ACKNOWLEDGMENTS This work has been supported by the ReGROUND project (http://reground.cs.kuleuven.be), which is a CHIST-ERA project funded by the EU H2020 framework program, the Research Foundation -Flanders  ... 
arXiv:1902.09937v1 fatcat:34la464m2jdaxpnl5p4csflzku

LCR-Net++: Multi-person 2D and 3D Pose Detection in Natural Images [article]

Gregory Rogez, Philippe Weinzaepfel, Cordelia Schmid
2019 IEEE Transactions on Software Engineering   accepted
We propose an end-to-end architecture for joint 2D and 3D human pose estimation in natural images.  ...  a classifier that scores the different pose proposals; and 3) a regressor that refines pose proposals both in 2D and 3D.  ...  Pose proposals integration LCR-Net outputs a set of refined pose proposals with multiple proposals covering each person present in the image.  ... 
doi:10.1109/tpami.2019.2892985 pmid:30640602 arXiv:1803.00455v3 fatcat:jslswqgavzcmjcwyp5lmiucsxq

Revisiting a single-stage method for face detection [article]

Nguyen Van Quang, Hiromasa Fujihara
2019 arXiv   pre-print
The design leverages the context information from the deeper layers in order to increase recall rate while maintaining a low false positive rate.  ...  In addition, we reduce the detection time by an improved inference procedure for decoding outputs faster. The inference time of a VGA (640×480) image was only approximately 26 ms with a Titan X GPU.  ...  Figure 2 (b) describes the detection module. 3) Anchor assigning strategy: In the training phase, we need to assign a ground truth box of faces to an anchor or a set of anchors.  ... 
arXiv:1902.01559v1 fatcat:wzccfvzlrbcnfax5fwleaowwdi

CNN-based Human Detection for UAVs in Search and Rescue [article]

Nikite Mesvan
2021 arXiv   pre-print
The model used in the research is a pre-trained model and is applied to test on a Raspberry Pi model B, which is attached on a Quadcopter, while a single camera is equipped at the bottom of the Quadcopter  ...  This paper proposes an approach for the first task of searching and detecting victims using a type of convolutional neural network technique, the Single Shot Detector (SSD) model, with the Quadcopter hardware  ...  The training mission is matching these anchor boxes with ground truth boxes.  ... 
arXiv:2110.01930v1 fatcat:u7zbqa3dajborbuskjpghkddwm

COOPERATIVE KNOWLEDGE BASED PERCEPTUAL ANCHORING

MARIOS DAOUTIS, SILVIA CORADESCHI, AMY LOUTFI
2012 International journal on artificial intelligence tools  
We present the results of the system and investigate different scenarios in which we use the commonsense together with perceptual knowledge, for communication, reasoning and exchange of information.  ...  In this work, we propose a model for connecting perceptual information to semantic information in a multi-agent setting.  ...  Also we thank Cycorp Inc. for offering their ResearchCyc platform used in this research. Finally we thank Sahar Asadi and Mathias Broxvall for their generous help with this research.  ... 
doi:10.1142/s0218213012500121 fatcat:e23fweofazfs3a6zgs4vampr34
« Previous Showing results 1 — 15 out of 23,474 results