3,406 Hits in 7.7 sec

Unsupervised Object Discovery and Segmentation in Videos

Samuel Schulter, Christian Leistner, Peter Roth, Horst Bischof
2013 Procedings of the British Machine Vision Conference 2013  
Furthermore, we demonstrate that the unsupervised learned appearance models also yield reasonable results for object detection on still images.  ...  Unsupervised object discovery is the task of finding recurring objects over an unsorted set of images without any human supervision, which becomes more and more important as the amount of visual data grows  ...  Given the motion segmentation, we extract tentative object proposals and discover semantically different object classes via a robust clustering approach.  ... 
doi:10.5244/c.27.53 dblp:conf/bmvc/SchulterLRB13 fatcat:ggrccv4n4beq5hx2zmu53vwxsm

Key-segments for video object segmentation

Yong Jae Lee, Jaechul Kim, Kristen Grauman
2011 2011 International Conference on Computer Vision  
We then compute a series of binary partitions among those candidate "key-segments" to discover hypothesis groups with persistent appearance and motion.  ...  We present an approach to discover and segment foreground object(s) in video.  ...  In video with a stationary background, moving foreground regions pop-out well with classic background subtraction algorithms (e.g., [28] ).  ... 
doi:10.1109/iccv.2011.6126471 dblp:conf/iccv/LeeKG11 fatcat:722kgevi5jhipdcad2m6nafz5q

Reusing 60GHz Radios for Mobile Radar Imaging

Yanzi Zhu, Yibo Zhu, Ben Y. Zhao, Haitao Zheng
2015 Proceedings of the 21st Annual International Conference on Mobile Computing and Networking - MobiCom '15  
The devices can also move while collecting reflection signals, creating a large synthetic aperture radar (SAR) for high-precision RF imaging.  ...  In addition to object location, our algorithm can discover a rich set of object surface properties at high precision, including object surface orientation, curvature, boundaries, and surface material.  ...  Background reflection from other objects can be handled via the multi-surface detection and imaging process described in §4.5.  ... 
doi:10.1145/2789168.2790112 dblp:conf/mobicom/ZhuZZZ15 fatcat:k2scrztbvjhznk4e4kuv67pwcm

Analyzing Wheels of Vehicles in Motion Using Laser Scanning

Andreas Mogelmose, Thomas B. Moeslund
2016 2016 IEEE Conference on Computer Vision and Pattern Recognition Workshops (CVPRW)  
Input from this is used to detect and analyze tires. A vertical lidar detects passing vehicles so individual tires can be combined into full vehicle models.  ...  Currently, no system is able to provide them with numbers on the wheel-widths of the vehicles using a particular stretch of road.  ...  The lack of detection performance comes down to the background model not being sufficiently robust.  ... 
doi:10.1109/cvprw.2016.199 dblp:conf/cvpr/MogelmoseM16 fatcat:zh4zso4on5ghlmy2igcb23dsyu

Tracking objects outside the line of sight using 2D intensity images

Jonathan Klein, Christoph Peters, Jaime Martín, Martin Laurenzis, Matthias B. Hullin
2016 Scientific Reports  
The observation of objects located in inaccessible regions is a recurring challenge in a wide variety of important applications.  ...  Recent work has shown that indirect diffuse light reflections can be used to reconstruct objects and two-dimensional (2D) patterns around a corner.  ...  We thank Frank Christnacher and Emmanuel Bacher for their help with the setup and discussion.  ... 
doi:10.1038/srep32491 pmid:27577969 pmcid:PMC5006175 fatcat:qewpml6rhnhjra7z3j4oeyrp7e

Sensor Fusion for Mobile Robot Navigation

M. Kam, Xiaoxun Zhu, P. Kalata
1997 Proceedings of the IEEE  
Integrating the sensor readings, the robot seeks to accomplish tasks such as constructing a map of its environment, locating itself in that map, and recognizing objects that should be avoided or sought  ...  It points to several further-research needs, including: robustness of decision rules; simultaneous consideration of self-location, motion planning, motion control and vehicle dynamics; the effect of sensor  ...  By triangulation, the tri-aural sensor is able to detect more than one object with one measurement.  ... 
doi:10.1109/jproc.1997.554212 fatcat:adofsedn2jd2bauazbslm46jea

Creating a virtual slide map from sputum smear images for region-of-interest localisation in automated microscopy

Bhavin Patel, Tania S. Douglas
2012 Computer Methods and Programs in Biomedicine  
The algorithm is inherently robust to changes in slide orientation and placement and showed high tolerance to illumination changes and robustness to noise.  ...  We use virtual slide maps together with geometric hashing to localise a query image, which then acts as the point of reference.  ...  If no possible matches with a significant number of votes are detected, the voting stage is repeated using another arbitrary basis pair in Q.  ... 
doi:10.1016/j.cmpb.2011.12.017 pmid:22257649 pmcid:PMC3350602 fatcat:cfagjuwxgzbh5mrry3smj4masu

Computational Methods for the Analysis of Footwear Impression Evidence [chapter]

Sargur N. Srihari, Yi Tang
2014 Studies in Computational Intelligence  
This paper begins with a comprehensive survey of existing methods, followed by identifying several gaps in technology.  ...  Retrieval performance of the proposed design with real crime scene images is evaluated and compared to that of previous methods.  ...  With clustering based on k recurring patterns as seed, 1000×k distance computations are needed; with k = 20, computation is reduced by 96%.  ... 
doi:10.1007/978-3-319-05885-6_15 fatcat:q7njd4iztrhvrgdmubx7erw4sa

Stages as Models of Scene Geometry

Vladimir Nedović, Arnold W M Smeulders, André Redert, Jan-Mark Geusebroek
2010 IEEE Transactions on Pattern Analysis and Machine Intelligence  
Consequently, we identify scene categorization as the first step towards robust and efficient depth estimation from single images.  ...  Stage information serves as the first approximation of global depth, narrowing down the search space in depth estimation and object localization.  ...  The objects come with almost infinite variation in appearance, as well as geometry. Scenes, on the other hand, show a much more regular pattern.  ... 
doi:10.1109/tpami.2009.174 pmid:20634560 fatcat:l53a2uj47raobpy6xmr6mtccpq

Visual Tracking: An Experimental Survey

2014 IEEE Transactions on Pattern Analysis and Machine Intelligence  
There is a large variety of trackers, which have been proposed in the literature during the last two decades with some mixed success.  ...  We find that in the evaluation practice the F-score is as effective as the object tracking accuracy (OTA) score.  ...  There are a few recurring patterns in motion models. The old uniform search has not yet lost its attractiveness due to its simplicity.  ... 
doi:10.1109/tpami.2013.230 pmid:26353314 fatcat:mw4q3amsfzdtjgzpooq2sjubpu

Learning Features by Watching Objects Move

Deepak Pathak, Ross Girshick, Piotr Dollar, Trevor Darrell, Bharath Hariharan
2017 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
When used for transfer learning on object detection, our representation significantly outperforms previous unsupervised approaches across multiple settings, especially when training data for the target  ...  Specifically, we use unsupervised motion-based segmentation on videos to obtain segments, which we use as 'pseudo ground truth' to train a convolutional network to segment objects from a single frame.  ...  move differently from pixels on the background.  ... 
doi:10.1109/cvpr.2017.638 dblp:conf/cvpr/PathakGDDH17 fatcat:teizzuwtkzbhrfiwm2zh4xoqde

Multiple structure recovery with maximum coverage

Luca Magri, Andrea Fusiello
2017 Machine Vision and Applications  
Two especially appealing characteristics of this method are the ease with which it can be implemented and its modularity with respect to the solver and to the sampling strategy.  ...  Few intelligible parameters need to be set and tuned, namely the inlier threshold and the number of desired models.  ...  We use the 51 real video sequences from the Hopkins 155 dataset [45] , each containing two or three moving objects, with no outliers.  ... 
doi:10.1007/s00138-017-0883-x fatcat:lb5dlgeiefht3fbnrj5g622roe

Audio Surveillance: a Systematic Review [article]

Marco Crocco, Marco Cristani, Andrea Trucco, Vittorio Murino
2014 arXiv   pre-print
subtraction, event classification, object tracking and situation analysis.  ...  To tackle this issue, audio sensory devices have been taken into account, both alone or in combination with video, giving birth, in the last decade, to a considerable amount of research.  ...  of hot objects against a colder background, such as people or moving vehicles [Dai et al., 2005] .  ... 
arXiv:1409.7787v1 fatcat:75lzk33wnnho5h3pbqr7nqaxxy

A study on video data mining

V. Vijayakumar, R. Nedunchezhian
2012 International Journal of Multimedia Information Retrieval  
Data mining is a process of extracting previously unknown knowledge and detecting the interesting patterns from a massive set of data.  ...  The objective of video data mining is to discover and describe interesting patterns from the huge amount of video data as it is one of the core problem areas of the data-mining research community.  ...  Zang and Klette [112] proposed an approach for extraction of a (new) moving object from the background and tracking of a moving object.  ... 
doi:10.1007/s13735-012-0016-2 fatcat:xuuf3w3b2rfcxlyevzndz6v62e

Cost of Quality in Crowdsourcing

Deniz Iren, Semih Bilgen
2014 Human Computation  
Crowdsourcing is a model which allows practitioners to access a relatively inexpensive and scalable workforce.  ...  Thus crowdsourcing practitioners have to rely heavily on certain quality assurance techniques to make sure that the end product complies with the quality requirements.  ...  Costs of errors surfaced after product delivery, non-detected errors yet to be found, non-conformances detected via quality assurance measures and rework performed to fix detected non-conformances are  ... 
doi:10.15346/hc.v1i2.14 dblp:journals/hc/IrenB14 fatcat:i5hoc3q5tze2tnqabsrwu4qzqi
« Previous Showing results 1 — 15 out of 3,406 results