instance learning
play

Instance Learning Boris Babenko, Ming-Hsuan Yang, Serge Belongie - PowerPoint PPT Presentation

Mar 06, 2024 •221 likes •921 views

Visual Tracking with Online Multiple Instance Learning Boris Babenko, Ming-Hsuan Yang, Serge Belongie Kelsie Zhao Content Goal Background: Tracking by Detection Previous Work New Tracking Solution MILTrack Online MILBoost

  1. Visual Tracking with Online Multiple Instance Learning Boris Babenko, Ming-Hsuan Yang, Serge Belongie Kelsie Zhao

  2. Content • Goal • Background: Tracking by Detection • Previous Work • New Tracking Solution • MILTrack • Online MILBoost • Experiments & Results

  3. Goal Track one arbitrary object in video, given its location in first frame

  4. Background: Tracking by detection • Frame 1 is labeled, tracker location known

  5. Background: Tracking by detection • Crop one positive and some negative patches near tracker x 1 x 3 x 2 Positive Negative

  6. Background: Tracking by detection • Use patches to train the classifier x 1 x 3 x 2 Positive Negative {(x 1 , 1), (x 2 , 0), (x 3 , 0)} Classifier

  7. Background: Tracking by detection • Frame 2 comes Classifier

  8. Background: Tracking by detection • Calculate classifier response within a range of the old tracker location X Old location Classifier

  9. Background: Tracking by detection • Find the maximum response location X X Old location New location Classifier

  10. Background: Tracking by detection • Move tracker Frame 1 Frame 2

  11. Background: Tracking by detection • Repeat Positive Negative Frame 2

  12. Background: Tracking by detection • Problem: If tracker location is not precise, might select bad training examples

  13. Background: Tracking by detection • Problem: If tracker location is not precise, might select bad training examples Model start to degrade!

  14. Background: Tracking by detection • Problem: If tracker location is not precise, might select bad training examples Model start to degrade! • How to select good training examples?

  15. Previous Work • Solution 1: multiple positive examples around tracker location x 2 x 5 x 4 x 3 x 1 {(x 1 , 1), (x 2 , 1), (x 3 , 1), (x 4 , 0), (x 5 , 0)} Classifier

  16. Previous Work • Solution 1 Might confuse classifier! x 2 x 5 x 4 x 3 x 1 {(x 1 , 1), (x 2 , 1), (x 3 , 1), (x 4 , 0), (x 5 , 0)} Classifier

  17. Previous Work • Solution 2: Multiple Instance Learning (MIL) [Keeler ‘90, Dietterich et al. ‘97]

  18. Previous Work: Multiple Instance Learning x 12 x 13 x 11 x 31 x 21 • Multiple examples in one bag [Keeler ‘90, Dietterich et al. ‘97]

  19. Previous Work: Multiple Instance Learning x 12 X 1 x 13 x 11 X 2 X 3 x 31 x 21 • Multiple examples in one bag [Keeler ‘90, Dietterich et al. ‘97]

  20. Previous Work: Multiple Instance Learning x 12 (X 1 , 1) x 13 x 11 (X 3 , 0) (X 2 , 0) x 31 x 21 • Multiple examples in one bag • One bag one label [Keeler ‘90, Dietterich et al. ‘97]

  21. Previous Work: Multiple Instance Learning x 12 (X 1 , 1) x 13 x 11 (X 3 , 0) (X 2 , 0) x 31 x 21 • Multiple examples in one bag • One bag one label • Bag Positive if at least one example is Positive [Keeler ‘90, Dietterich et al. ‘97]

  22. Previous Work: Multiple Instance Learning x 12 (X 1 , 1) x 13 x 11 (X 3 , 0) (X 2 , 0) x 31 x 21 {(X 1 , 1), (X 2 , 0), (X 3 , 0)} Classifier [Keeler ‘90, Dietterich et al. ‘97]

  23. Previous Work: Multiple Instance Learning MIL training input: 𝒀 𝟐 , 𝑧 1 … 𝒀 𝒐 , 𝑧 𝑜 , ( X 1 , 1) x 12 where, x 13 x 11 𝒀 𝒋 = 𝑦 𝑗1 … 𝑦 𝑗𝑛 , ( X 3 , 0) ( X 2 , 0) 𝑧 𝑗 = 𝑛𝑏𝑦 𝑘 𝑧 𝑗𝑘 x 31 x 21 [Keeler ‘90, Dietterich et al. ‘97]

  24. Previous Work: Multiple Instance Learning MIL training input: 𝒀 𝟐 , 𝑧 1 … 𝒀 𝒐 , 𝑧 𝑜 , ( X 1 , 1 ) x 12 where, x 13 x 11 𝒀 𝒋 = 𝑦 𝑗1 … 𝑦 𝑗𝑛 , ( X 3 , 0) ( X 2 , 0) 𝑧 𝑗 = 𝑛𝑏𝑦 𝑘 𝑧 𝑗𝑘 x 31 x 21 Bag babel is 1 if at least one instance is 1 [Keeler ‘90, Dietterich et al. ‘97]

  25. Now we have training examples! How to train the classifier?

  26. Previous Work: MILBoost • MIL + boosting Train a boosting classifier that maximizes log likelihood of bags 𝑚𝑝𝑕𝑀 = log (𝑞 𝑧 𝑗 𝒀 𝒋 )) 𝑗 where, 𝑞 𝑧 𝑗 𝒀 𝒋 ) = 1 − (1 − 𝑞 𝑧 𝑗 𝑦 𝑗𝑘 )) 𝑘 [Viola et al. ‘05]

  27. Previous Work: MILBoost • MIL + boosting Train a boosting classifier that maximizes log likelihood of bags 𝑚𝑝𝑕𝑀 = log (𝑞 𝑧 𝑗 𝒀 𝒋 )) 𝑗 where, 𝑞 𝑧 𝑗 𝒀 𝒋 ) = 1 − (1 − 𝑞 𝑧 𝑗 𝑦 𝑗𝑘 )) 𝑘 ~1 1 [Viola et al. ‘05]

  28. Previous Work: MILBoost • Problem: need all training examples [Viola et al. ‘05]

  29. Previous Work: MILBoost But in tracking, only current frame available [Viola et al. ‘05]

  30. Previous Work: MILBoost But in tracking, only current frame available Need an online training algorithm for MIL [Viola et al. ‘05]

  31. Main Contribution of this paper • Online-MILBoost: Online training for MIL-based classifier • MILTrack New tracking solution using Online-MILBoost

  32. MILTrack workflow X New frame comes in: 1. Crop out a set of image patches 𝑌 𝑡 = {𝑦| < 𝑡 1 𝑞𝑗𝑦𝑓𝑚𝑡 𝑔𝑠𝑝𝑛 𝑢𝑠𝑏𝑑𝑙𝑓𝑠 𝑚𝑝𝑑𝑏𝑢𝑗𝑝𝑜} 1: s = 35 in authors’ experiment Babenko et al., 09

  33. MILTrack workflow X New frame comes in: 1. Crop out a set of image patches 𝑌 𝑡 = {𝑦| < 𝑡 1 𝑞𝑗𝑦𝑓𝑚𝑡 𝑔𝑠𝑝𝑛 𝑢𝑠𝑏𝑑𝑙𝑓𝑠 𝑚𝑝𝑑𝑏𝑢𝑗𝑝𝑜} 1: s = 35 in authors’ experiment Babenko et al., 09

  34. MILTrack workflow X New frame comes in: 1. Crop out a set of image patches 𝑌 𝑡 = {𝑦| < 𝑡 1 𝑞𝑗𝑦𝑓𝑚𝑡 𝑔𝑠𝑝𝑛 𝑢𝑠𝑏𝑑𝑙𝑓𝑠 𝑚𝑝𝑑𝑏𝑢𝑗𝑝𝑜} 1: s = 35 in authors’ experiment Babenko et al., 09

  35. MILTrack workflow X New frame comes in: 1. Crop out a set of image patches 𝑌 𝑡 = {𝑦| < 𝑡 1 𝑞𝑗𝑦𝑓𝑚𝑡 𝑔𝑠𝑝𝑛 𝑢𝑠𝑏𝑑𝑙𝑓𝑠 𝑚𝑝𝑑𝑏𝑢𝑗𝑝𝑜} 1: s = 35 in authors’ experiment Babenko et al., 09

  36. MILTrack workflow New frame comes in: 1. Crop out a set of image patches 𝑌 𝑡 = {𝑦| < 𝑡 1 𝑞𝑗𝑦𝑓𝑚𝑡 𝑔𝑠𝑝𝑛 𝑢𝑠𝑏𝑑𝑙𝑓𝑠 𝑚𝑝𝑑𝑏𝑢𝑗𝑝𝑜} 1: s = 35 in authors’ experiment Babenko et al., 09

  37. MILTrack workflow New frame comes in: 1. Crop out a set of image patches 𝑌 𝑡 = {𝑦| < 𝑡 𝑞𝑗𝑦𝑓𝑚𝑡 𝑔𝑠𝑝𝑛 𝑢𝑠𝑏𝑑𝑙𝑓𝑠 𝑚𝑝𝑑𝑏𝑢𝑗𝑝𝑜} 2. Use MIL classifier to find new tracker location 𝑚 𝑜𝑓𝑥 = 𝑚(argmax 𝑦 ∈ 𝑌 𝑡 𝑞(𝑧 = 1|𝑦)) Babenko et al., 09

  38. MILTrack workflow New frame comes in: 1. Crop out a set of image patches 𝑌 𝑡 = {𝑦| < 𝑡 𝑞𝑗𝑦𝑓𝑚𝑡 𝑔𝑠𝑝𝑛 𝑢𝑠𝑏𝑑𝑙𝑓𝑠 𝑚𝑝𝑑𝑏𝑢𝑗𝑝𝑜} 2. Use MIL classifier to find new tracker location 𝑚 𝑜𝑓𝑥 = 𝑚(argmax 𝑦 ∈ 𝑌 𝑡 𝑞(𝑧 = 1|𝑦)) 3. 1) Crop positive examples 𝑌 𝑠 = {𝑦| < 𝑠 1 𝑞𝑗𝑦𝑓𝑚𝑡 𝑔𝑠𝑝𝑛 𝑢𝑠𝑏𝑑𝑙𝑓𝑠 𝑚𝑝𝑑𝑏𝑢𝑗𝑝𝑜} 1: r = 5 in authors’ experiment Babenko et al., 09

  39. MILTrack workflow New frame comes in: 1. Crop out a set of image patches 𝑌 𝑡 = {𝑦| < 𝑡 𝑞𝑗𝑦𝑓𝑚𝑡 𝑔𝑠𝑝𝑛 𝑢𝑠𝑏𝑑𝑙𝑓𝑠 𝑚𝑝𝑑𝑏𝑢𝑗𝑝𝑜} ( X 1 , 1) 2. Use MIL classifier to find new tracker location x 12 x 13 𝑚 𝑜𝑓𝑥 = 𝑚(argmax 𝑦 ∈ 𝑌 𝑡 𝑞(𝑧 = 1|𝑦)) x 11 3. 1) Crop positive examples 𝑌 𝑠 = {𝑦| < 𝑠 1 𝑞𝑗𝑦𝑓𝑚𝑡 𝑔𝑠𝑝𝑛 𝑢𝑠𝑏𝑑𝑙𝑓𝑠 𝑚𝑝𝑑𝑏𝑢𝑗𝑝𝑜} 1: r = 5 in authors’ experiment Babenko et al., 09

  40. MILTrack workflow New frame comes in: 1. Crop out a set of image patches 𝑌 𝑡 = {𝑦| < 𝑡 𝑞𝑗𝑦𝑓𝑚𝑡 𝑔𝑠𝑝𝑛 𝑢𝑠𝑏𝑑𝑙𝑓𝑠 𝑚𝑝𝑑𝑏𝑢𝑗𝑝𝑜} 2. Use MIL classifier to find new tracker location 𝑚 𝑜𝑓𝑥 = 𝑚(argmax 𝑦 ∈ 𝑌 𝑡 𝑞(𝑧 = 1|𝑦)) 3. 1) Crop positive examples 𝑌 𝑠 = {𝑦| < 𝑠 𝑞𝑗𝑦𝑓𝑚𝑡 𝑔𝑠𝑝𝑛 𝑢𝑠𝑏𝑑𝑙𝑓𝑠 𝑚𝑝𝑑𝑏𝑢𝑗𝑝𝑜} 3. 2) Crop Negative examples 𝑌 𝑠, 𝛾 = 𝑦 𝑠 𝑢𝑝 𝛾 1 𝑞𝑗𝑦𝑓𝑚𝑡 𝑏𝑥𝑏𝑧 𝑔𝑠𝑝𝑛 𝑢𝑠𝑏𝑑𝑙𝑓𝑠 𝑚𝑝𝑑𝑏𝑢𝑗𝑝𝑜} 1: 𝛾 = 50 in authors’ experiment Babenko et al., 09

  41. MILTrack workflow New frame comes in: 1. Crop out a set of image patches 𝑌 𝑡 = {𝑦| < 𝑡 𝑞𝑗𝑦𝑓𝑚𝑡 𝑔𝑠𝑝𝑛 𝑢𝑠𝑏𝑑𝑙𝑓𝑠 𝑚𝑝𝑑𝑏𝑢𝑗𝑝𝑜} 2. Use MIL classifier to find new tracker location 𝑚 𝑜𝑓𝑥 = 𝑚(argmax 𝑦 ∈ 𝑌 𝑡 𝑞(𝑧 = 1|𝑦)) x 31 x 21 3. 1) Crop positive examples 𝑌 𝑠 = {𝑦| < 𝑠 𝑞𝑗𝑦𝑓𝑚𝑡 𝑔𝑠𝑝𝑛 𝑢𝑠𝑏𝑑𝑙𝑓𝑠 𝑚𝑝𝑑𝑏𝑢𝑗𝑝𝑜} ( X 3 , 0) ( X 2 , 0) 3. 2) Crop Negative examples 𝑌 𝑠, 𝛾 = 𝑦 𝑠 𝑢𝑝 𝛾 1 𝑞𝑗𝑦𝑓𝑚𝑡 𝑏𝑥𝑏𝑧 𝑔𝑠𝑝𝑛 𝑢𝑠𝑏𝑑𝑙𝑓𝑠 𝑚𝑝𝑑𝑏𝑢𝑗𝑝𝑜} 1: 𝛾 = 50 in authors’ experiment Babenko et al., 09

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

Learning for Categorization Sample Category Learning Problem A training example is an instance

Learning for Categorization Sample Category Learning Problem A training example is an instance

Categorization (review) Given: A description of an instance, x X , where X is the instance language or instance space . Text Categorization A fixed set of categories: C= { c 1 , c 2 , c n } Determine: The category of x

151 views • 4 slides
INSTANCE BASED LEARNING 2 Instance-Based Learning Distance function defines whats learned

INSTANCE BASED LEARNING 2 Instance-Based Learning Distance function defines whats learned

LEARNING FROM OBSERVATIONS INSTANCE BASED LEARNING 2 Instance-Based Learning Distance function defines whats learned Most instance-based schemes use Euclidean distance : a (1) and a (2) : two instances with k attributes Taking

705 views • 41 slides
1 Hypothesis Space Inductive Learning Hypothesis Restrict learned functions a priori to a

1 Hypothesis Space Inductive Learning Hypothesis Restrict learned functions a priori to a

Classification (Categorization) Given: A description of an instance, x X , where X is the CS 391L: Machine Learning: instance language or instance space . A fixed set of categories: C= { c 1 , c 2 , c n } Inductive

559 views • 9 slides
Classify as a Whole? MULTIPLE INSTANCE LEARNING Set Learning? Multi-Set Learning? Marco Loog

Classify as a Whole? MULTIPLE INSTANCE LEARNING Set Learning? Multi-Set Learning? Marco Loog

TUDelft PRLab NL Classify as a Whole? MULTIPLE INSTANCE LEARNING Set Learning? Multi-Set Learning? Marco Loog Pattern Recognition Laboratory Delft University of Technology heavily inspired by slides from Veronika Cheplygina and David MJ

486 views • 32 slides
Instance Based Learning k -Nearest Neighbor Locally weighted regression Radial basis

Instance Based Learning k -Nearest Neighbor Locally weighted regression Radial basis

Instance Based Learning k -Nearest Neighbor Locally weighted regression Radial basis functions Case-based reasoning Lazy and eager learning 1 Instance-Based Learning Key idea: just store all training examples x i , f ( x i

316 views • 16 slides
Explaining the Stars: Weighted Multiple-Instance Learning for Aspect-Based Sentiment Analysis

Explaining the Stars: Weighted Multiple-Instance Learning for Aspect-Based Sentiment Analysis

Motivation Multiple-instance learning The proposed model Experiments Conclusion Explaining the Stars: Weighted Multiple-Instance Learning for Aspect-Based Sentiment Analysis Nikolaos Pappas and Andrei Popescu-Belis Idiap Research Institute,

499 views • 23 slides
Nearest Neighbor Learning (Instance Based Learning) l Classify based on local similarity l Ranges

Nearest Neighbor Learning (Instance Based Learning) l Classify based on local similarity l Ranges

Nearest Neighbor Learning (Instance Based Learning) l Classify based on local similarity l Ranges from simple nearest neighbor to case-based and analogical reasoning l Use local information near the current query instance to decide the

490 views • 33 slides
Instance-based Learning Hamid R. Rabiee Spring 2015 http://ce.sharif.edu/courses/93-94/2/ce717-1

Instance-based Learning Hamid R. Rabiee Spring 2015 http://ce.sharif.edu/courses/93-94/2/ce717-1

Instance-based Learning Hamid R. Rabiee Spring 2015 http://ce.sharif.edu/courses/93-94/2/ce717-1 Agenda Agenda Instance-based learning K-Nearest Neighbor Locally-weighted Regression Radial Basis Function Networks Case-based

475 views • 30 slides
A meta-learning system for multi-instance classification Gitte Vanwinckelen and Hendrik Blockeel

A meta-learning system for multi-instance classification Gitte Vanwinckelen and Hendrik Blockeel

A meta-learning system for multi-instance classification Gitte Vanwinckelen and Hendrik Blockeel KU Leuven, Belgium Motivation Performed extensive evaluation of multi-instance (MI) learners on datasets from different domains

330 views • 19 slides
Learning Object Bounding Boxes for 3D Instance Segmentation on Point Clouds B. Yang, J. Wang,

Learning Object Bounding Boxes for 3D Instance Segmentation on Point Clouds B. Yang, J. Wang,

Neural Information Processing Systems (NeurIPS) 2019 Learning Object Bounding Boxes for 3D Instance Segmentation on Point Clouds B. Yang, J. Wang, R. Clark, Q. Hu, S. Wang, A. Markham, N. Trigoni Background: 2D Instance Segmentation 3D

345 views • 11 slides
Introduction to Machine Learning 3. Instance Based Learning Alex Smola Carnegie Mellon

Introduction to Machine Learning 3. Instance Based Learning Alex Smola Carnegie Mellon

Introduction to Machine Learning 3. Instance Based Learning Alex Smola Carnegie Mellon University http://alex.smola.org/teaching/cmu2013-10-701 10-701 Outline Parzen Windows Kernels, algorithm Model selection Crossvalidation, leave

1.1k views • 68 slides
Instance-based Learning CE-717: Machine Learning Sharif University of Technology M. Soleymani

Instance-based Learning CE-717: Machine Learning Sharif University of Technology M. Soleymani

Instance-based Learning CE-717: Machine Learning Sharif University of Technology M. Soleymani Fall 2018 Outline } Non-parametric approach } Unsupervised: Non-parametric density estimation } Parzen Windows } Kn-Nearest Neighbor Density

863 views • 52 slides
Instance Based Learning [Read Ch. 8] k -Nearest Neigh b or Lo cally w eigh

Instance Based Learning [Read Ch. 8] k -Nearest Neigh b or Lo cally w eigh

Instance Based Learning [Read Ch. 8] k -Nearest Neigh b or Lo cally w eigh ted regression Radial basis functions Case-based reasoning Lazy and eager learning 199 lecture slides for textb o ok

540 views • 16 slides
Multiple Instance Learning for Fast, Stable and Early RNN Predictions Don Dennis , Microsoft

Multiple Instance Learning for Fast, Stable and Early RNN Predictions Don Dennis , Microsoft

The Edge of Machine Learning Multiple Instance Learning for Fast, Stable and Early RNN Predictions Don Dennis , Microsoft Research India, Joint work with Chirag P., Harsha and Prateek Accepted to NIPS 18 1 Algorithms for the IDE - EdgeML

977 views • 70 slides
About any instance (fi rst instance, appeal, cassation, the ARTYUSHENKO &amp; PARTNERS IS THE

About any instance (fi rst instance, appeal, cassation, the ARTYUSHENKO &amp; PARTNERS IS THE

eff ective way we will use all remedies Legal opinions on specifi c issues About any instance (fi rst instance, appeal, cassation, the ARTYUSHENKO &amp; PARTNERS IS THE ONLY REAL-ESTATE AND CONSTRUCTION BOUTIQUE LAW FIRM IN KAZAKHSTAN.

467 views • 8 slides
Instance Based Learning Based on Machine Learning, T. Mitchell, McGRAW Hill, 1997, ch. 8

Instance Based Learning Based on Machine Learning, T. Mitchell, McGRAW Hill, 1997, ch. 8

0. Instance Based Learning Based on Machine Learning, T. Mitchell, McGRAW Hill, 1997, ch. 8 Acknowledgement: The present slides are an adaptation of slides drawn by T. Mitchell 1. Key ideas: training: simply store all training examples

598 views • 17 slides
Resou ource Allocat ation w with th a a Bu Budget et Constraint int f for C Comput

Resou ource Allocat ation w with th a a Bu Budget et Constraint int f for C Comput

Resou ource Allocat ation w with th a a Bu Budget et Constraint int f for C Comput uting ing Independe ndent nt T as asks i in the the Cl Clou oud Weiming Shi and Bo Hong School of Electrical and Computer Engineering Georgia

331 views • 30 slides
Expected and Actual Forecast Errors by a Non-normal Model of El Nio Ccile Penland

Expected and Actual Forecast Errors by a Non-normal Model of El Nio Ccile Penland

Expected and Actual Forecast Errors by a Non-normal Model of El Nio Ccile Penland NOAA/ESRL/Physical Sciences Division Outline Some phenomenology: What is El Nio The Annual Cycle Deviations from it, especially El Nio An

533 views • 34 slides
HCAL TriDAS Status Drew Baden, University of Maryland For the HCAL Group: Boston University

HCAL TriDAS Status Drew Baden, University of Maryland For the HCAL Group: Boston University

HCAL TriDAS Status Drew Baden, University of Maryland For the HCAL Group: Boston University Fermilab Princeton University University Maryland HCAL TriDAS 1 21-Jun-2005 Overview S-Link: 64 bits @ 25 MHz Trigger Primitives

687 views • 34 slides
- masses M , stiffness K - modes Physical linear dynamic model. Natural excitation : - swell

- masses M , stiffness K - modes Physical linear dynamic model. Natural excitation : - swell

Statistical Approaches Introduction to Industrial Monitoring Problems - Signal processing and on-board monitoring Fault Detection and Isolation Early detection of slight deviations M. Basseville, Q. Zhang, A. Benveniste with respect to a

1.31k views • 13 slides
Kanade-Lucas-Tomasi (KLT) Tracker 16-385 Computer Vision (Kris Kitani) Carnegie Mellon University

Kanade-Lucas-Tomasi (KLT) Tracker 16-385 Computer Vision (Kris Kitani) Carnegie Mellon University

Kanade-Lucas-Tomasi (KLT) Tracker 16-385 Computer Vision (Kris Kitani) Carnegie Mellon University https://www.youtube.com/watch?v=rwIjkECpY0M Feature-based tracking Up to now, weve been aligning entire images but we can also track just

348 views • 24 slides
CLAS12 Tracking Overview and Progress Veronique Ziegler First Experiment Workshop CLAS 12 First

CLAS12 Tracking Overview and Progress Veronique Ziegler First Experiment Workshop CLAS 12 First

CLAS12 Tracking Overview and Progress Veronique Ziegler First Experiment Workshop CLAS 12 First Experiment Workshop 1 Central Vertex Tracker Reconstruction New package clasrec-CVT contains algorithms to reconstruct events using BMT and

309 views • 27 slides
Scintillation detector based on InAs quantum dots in a GaAs semiconductor matrix for charged

Scintillation detector based on InAs quantum dots in a GaAs semiconductor matrix for charged

Scintillation detector based on InAs quantum dots in a GaAs semiconductor matrix for charged particle tracking or can one build a tracker out of scintillating wafers ? S.Oktyabrsky 1 , M.Yakimov 1 , V.Tokranov 1 , K.Dropiewsky 1 , A.Minns 1 ,

491 views • 18 slides
Distributed Online Tracking Mingwang Tang, Feifei Li , Yufei Tao 1 Motivation and challenge

Distributed Online Tracking Mingwang Tang, Feifei Li , Yufei Tao 1 Motivation and challenge

Distributed Online Tracking Mingwang Tang, Feifei Li , Yufei Tao 1 Motivation and challenge Sensor network LBS service 2 Motivation and challenge Sensor network LBS service Observation: Large distributed data are being generated

1.19k views • 67 slides

More recommend