Real Time Object Tracking

Download Real Time Object Tracking

Post on 21-Jan-2015

10.995 views

Category:

Technology

3 download

Embed Size (px)

DESCRIPTION

Some discussion about real time object tracking and detection methods.

TRANSCRIPT

  • 1. REAL-TIME OBJECT DETECTIONAND TRACKING By: Vanya V. Valindria Hammad Naeem Rui Hua
  • 2. Outline Introduction Hardware in RT Object Detetion & Tracking Methods Traditional Methods: Modern Methods: Absolute Differences Census Method KLT Feature Based Method Meanshift Result and Conclusion
  • 3. IntroductionDefinition:Object detection detect a particular object in an imageObject tracking to track an object(or multiple objects)over a sequence ofimages
  • 4. Application: Traffic Informationhttp://www.youtube.com/watch?v=vA35sXbn7zs
  • 5. Application: Surveillancehttp://www.youtube.com/watch?v=o25fClk9cdg
  • 6. Application: Mobile Robothttp://www.youtube.com/watch?v=Q4zycRGJFFs
  • 7. Problems?? Temporal variation/dynamic environment Abrupt object or camera motion Multi-camera? Multi-objects? Computational expensive
  • 8. Hardware in Real-time Tracking MEMORYImportant Tracking system encountering limited memory problems. FRAME RATE ~30 FPS PROCESSORS - DSP Allow saturated arithmetic operation Powerful operation ability Can do several memory accesses in a single instruction
  • 9. Object Detection and Tracking In a video sequence an object is said to be in motion, if it is changing its location with respect to its background The motion tracking is actually the process of keeping tracks of that moving object in video sequence i.e. position of moving object at certain time etc.
  • 10. Flow Chart Idle Image acquisition Object Detection Image acquisition Object tracking Object No Lost? Y es
  • 11. Method 1: Absolute Differences= Image subtraction D(t)=I(ti) I(tj)Gives an image frame with changed and unchanged regionsIdeal Case for no motion: I(ti) = I(tj), D(t)=0
  • 12. Movingobjectsaredetected
  • 13. Results: Frame1 Frame10 Difference of Two Frames
  • 14. Absolute DifferenceMethods for Motion Detection Frame Differencing Background SubtractionDraw Backs:involves a lot of computations Not feasible for DSP implementation
  • 15. Method 2: Census Transforms124 74 32 1 1 0124 64 18 If (Center pixel < Neighbor 1 x 0 pixel) Neighbor pixel = 1157 116 84 1 1 1 Signature Vector Generation 1 1 0 1 x 0 Signature Vector 11011101 1 1 1
  • 16. Image128 26 125 243 87 Signature Vectors96 76 43 236 125 10110101 00101011128 129 235 229 209 Signatur vector generation for . all pixels228 251 229 221 234 . .227 221 35 58 98 10111010 List 10110101 Generation List 00101011 population . . . Generated List 10111010 Signature vector matching
  • 17. Census Transform: Advantages: Compare only two values 0 or 1. Similar Illumination Variation for pixel and neighbouring pixels Draw Backs: As we only deal with only 0`s and 1`s, this method is sensitive to noise. Calculate, store and match process computationally Expensive
  • 18. Method 3: Morphology Based Object TrackingBackground Frame Object estimation differencing Registration
  • 19. Morphology Based Object Tracking Image DifferencingBackground ThresholdingEstimation Contours are registered Object Width, height and histogram are recorded for each contourRegistration Each object represented by a feature vector (the length, Feature width, area and histogram of the object) Vector
  • 20. Segmented object Tracked object
  • 21. Morphology Based TechniquesAdvantages:Can Track Multiple objects Objects are registered based on their anatomy Helpful for Object MergingDraw Backs:Object registration complex and slow process For multiple object registration per frame more complex
  • 22. Method 4: Lucas-Kanade Technique Visual motion pattern of objects and surface in a scene by Optical Flow Frame 1 Frame 2
  • 23. Method 5: Mean shift An algorithm that iteratively shifts a data point to the average of data points in its neighborhood Choose a search window Compute the MEAN size in the initial location location in the search window Repeat until Center the search convergence window at the mean
  • 24. Intuitive Description Region of interest Center of mass Mean Shift vector Objective : Find the densest region Distribution of identical balls
  • 25. Intuitive Description Region of interest Center of mass Mean Shift vector Objective : Find the densest region Distribution of identical balls
  • 26. Intuitive Description Region of interest Center of mass Mean Shift vector Objective : Find the densest region Distribution of identical balls
  • 27. Intuitive Description Region of interest Center of mass Mean Shift vector Objective : Find the densest region Distribution of identical balls
  • 28. Intuitive Description Region of interest Center of mass Mean Shift vector Objective : Find the densest region Distribution of identical balls
  • 29. Intuitive Description Region of interest Center of mass Mean Shift vector Objective : Find the densest region Distribution of identical balls
  • 30. Intuitive Description Region of interest Center of mass Objective : Find the densest region Distribution of identical balls
  • 31. Process
  • 32. CAMSHIFT--Continously Adaptive MeanshiftModified to adapt dynamically to the colour probability distributionsMore real time For each frame-> MEAN-SHIFT is applied with several iteration Store the location of the mean and calculate new window size for next frame
  • 33. New development Combine with different features. SIFT features, colour feature & texture information Camshift algorithm combined with the Kalman filter.
  • 34. Result Arithmetic and Time taken Algorithm Logic by operations Algorithm Absolute 4230100 16 Differencing Census 2416000 5. 4 Transform Morphological 352210 14.2 Tracking Kanade Lucas 500825 0.486
  • 35. Comparison Computationally Easy to implement expensive Absolute Differences Allows continuous Slow and low tracking accuracy Computationally...

Recommended

View more >