Search
Search Funnelback University
- Refined by:
- Date: 2014
1 -
40 of
40
search results for TALK:PC53 20 |u:mi.eng.cam.ac.uk
where 0
match all words and 40
match some words.
Results that match 1 of 2 words
-
Oliver Williams University of Cambridge Andrew Blake Microsoft…
mi.eng.cam.ac.uk/~cipolla/archive/Presentations/2003-ICCV-tracking-RVM.pdf19 May 2014: Treat score as function of. position. -100 -80 -60 -40 -20 0 20 40 60 80 100-3. ... 100 Training. examples. 12 Relevant vectors. remain. 0 10 20 30 40 50 60 70 80 90 100-0.6. -
slides.dvi
mi.eng.cam.ac.uk/~mjfg/Bilbao14/talk.pdf25 Jun 2014: Cambridge University. Engineering DepartmenteNTERFACE June 2014 20. Controllable and Adaptable Statistical Parametric Speech Synthesis Systems. ... recognition in different environments [19]/speaker identification [20]. Cambridge University. Engineering -
Knill_IS14_1.dvi
mi.eng.cam.ac.uk/~ar527/knill_is2014a.pdf10 Nov 2014: from multiple languages [20]. Context dependent (CD) out-. put layer targets were adopted as they have been found to yield. ... 307–312. [20] G. Hinton, L. Deng et al., “Deep Neural Networks for AcousticModeling in Speech Recognition,” Signal -
How to prepare and deliver a presentation
mi.eng.cam.ac.uk/~cipolla/archive/Presentations/2007-Making-Presentations.pdf19 May 2014: 10 20 30 40 50 60. See. Hear. Fear of public speaking. • ... Arial and 32 pt. Arial and 28 pt. Arial and 20 pt. -
INFINITE STRUCTURED SUPPORT VECTOR MACHINES FOR SPEECH RECOGNITION…
mi.eng.cam.ac.uk/~mjfg/yang_ICASSP14.pdf28 Apr 2014: Avgtesta testb testc. HMM — 9.83 9.11 9.53 9.48SVM. Log-Like8.29 7.90 8.61 8.20. ... 20, no. 7, pp. 2149–2158, 2012. [5] Robert A. Jacobs, Michael I. -
Combining Tandem and Hybrid Systems for Improved Speech…
mi.eng.cam.ac.uk/~ar527/rath_is2014a.pdf10 Nov 2014: The underlying context-dependent states were specifiedusing state [19, 20, 21], rather than phone-state, roots of thedecision tree. ... 20] S. J. Young, J. J. Odell, and P. C. Woodland, “Tree-basedstate tying for high accuracy acoustic modelling,” in -
Model-Based Hand Tracking
mi.eng.cam.ac.uk/~cipolla/archive/Presentations/2003-MOSX-Hand-Detection.pdf19 May 2014: 9.0 20.0 17.0. Pruning the Search-Tree. 9.0 20.0 17.0. Pruning the Search-Tree. ... 9.0 20.0 17.0. 10.0 4.5 6.7. Pruning the Search-Tree. 9.0 20.0 17.0. -
ICASSP2014b.dvi
mi.eng.cam.ac.uk/~mjfg/yoshioka_ICASSP14.pdf3 Apr 2014: Audio, Speech, Language Process.,vol. 20, no. 1, pp. 30–42, 2012. [2] L. ... Int. Conf.Acoust., Speech, Signal Process., 2013, pp. 7063–7067. [20] M. J. -
Rank-HCI
mi.eng.cam.ac.uk/~cipolla/archive/Presentations/2005-Rank-HCI.pdf19 May 2014: Detecting frontal faces. 0 10 20 30 40 50 60 70 80. -
EFFICIENT LATTICE RESCORING USINGRECURRENT NEURAL NETWORK LANGUAGE…
mi.eng.cam.ac.uk/~mjfg/xl207_ICASSP14a.pdf28 Apr 2014: 3. HISTORY CONTEXT CLUSTERING FOR RNNLMS. Efficient use of language models in speech recognizers [20, 19, 17]requires that the context dependent states representing different his-tories during search can be ... In common with the n-gram history based -
Face and hand detection - ICT
mi.eng.cam.ac.uk/~cipolla/archive/Presentations/2003-ICT-Hands-Faces.pdf19 May 2014: Detecting frontal faces. 0 10 20 30 40 50 60 70 80. -
Cambridge University Engineering Department: News item
mi.eng.cam.ac.uk/~cipolla/archive/Public-Understanding/2012-CUED-Metail.pdf7 Nov 2014: 20/03/2012 09:31Cambridge University Engineering Department: News item. Page 1 of 4http://www.eng.cam.ac.uk/news/stories/2012/Me_tail/. ... Today, fashion e-commerceis one of the fastest growing Internet sectorswith a UK growth rate of around 20% perannum -
2010-FREng
mi.eng.cam.ac.uk/~cipolla/archive/Public-Understanding/2010-CUED-FREng.pdf7 Nov 2014: 20 September 2010. Congratulations to Roberto Cipolla Professor ofInformation Engineering and Dr Ivor Day SeniorRolls-Royce Research Fellow, at the WhittleLaboratory, who have been elected as Fellows of theRoyal Academy of -
3-D Tech Lets Online Shoppers 'Try on' Clothes |…
mi.eng.cam.ac.uk/~cipolla/archive/Public-Understanding/2013-Entrepreneur-Metail.pdf7 Nov 2014: This year Adey-oola hopes to secure customers in the U.S., but says he's going after only the top 20 e-tailers. -
Combining Tandem and Hybrid Systems for Improved Speech…
mi.eng.cam.ac.uk/~mjfg/interspeech14-rath.pdf23 Jun 2014: The underlying context-dependent states were specifiedusing state [19, 20, 21], rather than phone-state, roots of thedecision tree. ... 20] S. J. Young, J. J. Odell, and P. C. Woodland, “Tree-basedstate tying for high accuracy acoustic modelling,” in -
Knill_IS14_final.dvi
mi.eng.cam.ac.uk/~mjfg/interspeech14-knill.pdf23 Jun 2014: ing [20] state position root phonetic decision trees [21, 22] are. ... 25, pp. 519–534, 2010. [20] S. Young, J. Odell, and P. -
Cambridge University Engineering Department: News item
mi.eng.cam.ac.uk/~cipolla/archive/Public-Understanding/2011-CUED-Kinect.pdf7 Nov 2014: 20 January 2011. Dr Jamie Shotton completed his PhD in the MachineIntelligence Lab with Professor Cipolla, here at theDepartment of Engineering from 2003-2007. -
3D computer model emerges from a pile of snaps - ProQuest
mi.eng.cam.ac.uk/~cipolla/archive/Public-Understanding/2005-NewScientist-3D-models.pdf7 Nov 2014: Copyright Reed Business Information UK Aug 20-Aug 26, 2005. -
unsup.eps
mi.eng.cam.ac.uk/~ar527/ragni_is2014a.pdf20 Jun 2014: When applied to speakers, the SFM essentiallyyields a simplified voice morphing [20] scheme. ... 20] H. Ye and S. J. Young, “High quality voice morphing,” inICASSP,vol. -
unsup.eps
mi.eng.cam.ac.uk/~mjfg/interspeech14-ragni.pdf23 Jun 2014: When applied to speakers, the SFM essentiallyyields a simplified voice morphing [20] scheme. ... 20] H. Ye and S. J. Young, “High quality voice morphing,” inICASSP,vol. -
SPEECH RECOGNITION AND KEYWORD SPOTTING FOR LOW RESOURCELANGUAGES:…
mi.eng.cam.ac.uk/~mjfg/sltu14_mjfg.pdf6 Jun 2014: An alternative ap-proach is to make use of data from other languages [6, 19, 20, 21],multi-language resources. ... example [20]. Thus the normalisation summation (shown in blue)acts on single language. -
Photo recognition software gives location - 10 April 2004 - New…
mi.eng.cam.ac.uk/~cipolla/archive/Public-Understanding/2004-NewScientist-Photo-recognition.pdf7 Nov 2014: 10/09/2008 11:02Photo recognition software gives location - 10 April 2004 - New Scientist. Page 1 of 3file:///Users/cipolla/Desktop/Public-Understanding-Archive/2004-Photorecognition-NewScientist.webarchive. SEARCH Tips. New Scientist | Space | -
D:/docs/reports/tech_17/tech_17.dvi
mi.eng.cam.ac.uk/reports/svr-ftp/treece_tr691.pdf4 Aug 2014: 3030. 30. 2020 20. 10 1010. 10 1010. 20. 20. 20. ... 10. 10. 20. 20. 30. 30. 40. 40506070. 80. thickness (mm). -
SPEECH RECOGNITION AND KEYWORD SPOTTING FOR LOW RESOURCELANGUAGES:…
mi.eng.cam.ac.uk/~ar527/gales_sltu2014a.pdf10 Nov 2014: An alternative ap-proach is to make use of data from other languages [6, 19, 20, 21],multi-language resources. ... example [20]. Thus the normalisation summation (shown in blue)acts on single language. -
RedHerring.com -- The Business of Technology
mi.eng.cam.ac.uk/~cipolla/archive/Public-Understanding/2000-RedHerring-Visual-Tracking-Robotics.pdf7 Nov 2014: According to Mr.Gordon, the 4DI system can complete its measurements in 20 seconds; the same task would take an employee using a handheld measuring devicemore than 30 minutes. -
The EarthThe faults that move duringearthquakes do not always ...
mi.eng.cam.ac.uk/~cipolla/archive/Public-Understanding/2007-CAM-vision.pdf7 Nov 2014: Top. Map view of the central Iraniandesert with grid crosses 20 km apart.After an earthquake in 1998, a radarinterferogram was draped over thetopography so that each colour fringedistinguished a change -
2012 ICARV invited
mi.eng.cam.ac.uk/~cipolla/archive/Presentations/2014-Cipolla-Overview.pdf19 May 2014: Vector. 3 Machine Learning. 20. Training data – supervised learning. 21. ... 04/15 15-02-13. 05/15 18-02-13. 06/15 20-02-13. 07/15 22-02-13. 08/15 25-02-13. 09/15 04-03-13. -
With Zappar, Augmented Reality Is Ready for Prime Time | TIME.com
mi.eng.cam.ac.uk/~cipolla/archive/Public-Understanding/2012-TIME-Zappar.pdf7 Nov 2014: 20 Quick Chicken RecipesiVillage. Three Ways to Extend theBattery Life of Smartphonesand TabletsEntrepreneur. -
Noname manuscript No.(will be inserted by the editor) Using ...
mi.eng.cam.ac.uk/~cipolla/publications/article/2013-IJCV-dense-AAM.pdf19 May 2014: entropy of the set of transformed images [20]. These. allowable transformations are general in that they can. ... 1 4 8 12 16 20 240.2. 0.4. 0.6. 0.8. 1. -
Exposure, built in the Netherlands, took Scottish pylon-makers and ...
mi.eng.cam.ac.uk/~cipolla/archive/Public-Understanding/2010-Guardian-Gormley-Exposure.pdf7 Nov 2014: Exposure, built in the Netherlands, took Scottish pylon-makers and Dutch engineerssix years to complete. (115)Tweet this (62). Antony Gormley drops 60-tonne load for monumentalsculpture. Dutch residents have nicknamed Gormley's new work as 'the -
Noname manuscript No.(will be inserted by the editor) Using ...
mi.eng.cam.ac.uk/~cipolla/publications/inproceedings/2013-Anderson-IJCV.pdf19 May 2014: 0 20 40 600. 0.2. 0.4. 0.6. 0.8. 1. Sorted edge index. ... base flow. 1 4 8 12 16 20 240.2. 0.4. 0.6. -
東芝レビュー2007年12月
mi.eng.cam.ac.uk/~cipolla/archive/Public-Understanding/2007-Toshiba-Review.pdf7 Nov 2014: This is done by initialising a colour model using a boosted hand detector [19, 20]. ... Intell., 2008. to appear. [20] B. Stenger. Template-based hand pose recognition using multiple cues. -
Cambridge University And Toshiba Bring A New Dimension To Photos |…
mi.eng.cam.ac.uk/~cipolla/archive/Public-Understanding/2011-Meerkats-3D-infrastructure.pdf7 Nov 2014: Page 2 of 2http://holdsworthnewswire.wordpress.com/2011/12/17/cambridge-university-and-toshiba-bring-a-new-dimension-to-photos/. 20 Years of Toshiba Research in Cambridge. -
Int J Comput VisDOI 10.1007/s11263-013-0673-5 Mixture of Trees…
mi.eng.cam.ac.uk/~cipolla/archive/Publications/article/2013-IJCV-MOT-video-segmentation.pdf19 May 2014: Penguin 20,028 42 6,627 1,705 1,367 19,310 1,212 3,218 1,736 29,461. -
Noname manuscript No.(will be inserted by the editor) Using ...
mi.eng.cam.ac.uk/~cipolla/archive/Publications/article/2013-IJCV-dense-AAM.pdf19 May 2014: entropy of the set of transformed images [20]. These. allowable transformations are general in that they can. ... 1 4 8 12 16 20 240.2. 0.4. 0.6. 0.8. 1. -
Do the maths - Times Online
mi.eng.cam.ac.uk/~cipolla/archive/Public-Understanding/2009-Sunday-Times-Maths.pdf7 Nov 2014: There’s alwaysa uniquely bright kid in any school,” he says. “But I’m interestedin the 20% doing maths and science to A-level and beyond.When I was in the ... even for the 20% of school leavers who want to doscience and maths at university. -
Noname manuscript No.(will be inserted by the editor) Using ...
mi.eng.cam.ac.uk/~cipolla/archive/Publications/inproceedings/2013-Anderson-IJCV.pdf19 May 2014: 0 20 40 600. 0.2. 0.4. 0.6. 0.8. 1. Sorted edge index. ... base flow. 1 4 8 12 16 20 240.2. 0.4. 0.6. -
Discriminative models for speech recognition Anton Ragni Peterhouse…
mi.eng.cam.ac.uk/~mjfg/thesis_ar527.pdf2 Apr 2014: trained (VAT) and discriminative VAT (DVAT) HMM WER per-. formance on the Aurora 2 task averaged over 0-20 dB. ... q̂1:T = arg maxq1:T. {aq0,q1. Tt=1. bqt (ot)aqt,qt1. }(2.20). A formal technique, known as Viterbi algorithm [188], is commonly used to -
newsletter.indd
mi.eng.cam.ac.uk/~cipolla/archive/Public-Understanding/2005-Insight-Tracking-Crowds.pdf7 Nov 2014: It would not only last much longer than a battery of the same weight (about 20 times longer at 10 per cent effi ciency), but would also require little time to ... Alongside Research Services Division, the Mission Tour met with over 20 University -
Int J Comput VisDOI 10.1007/s11263-013-0673-5 Mixture of Trees…
mi.eng.cam.ac.uk/~cipolla/publications/article/2013-IJCV-MOT-video-segmentation.pdf19 May 2014: Penguin 20,028 42 6,627 1,705 1,367 19,310 1,212 3,218 1,736 29,461.
Search history
Recently clicked results
Recently clicked results
Your click history is empty.
Recent searches
Recent searches
Your search history is empty.