340 likes | 528 Views
Tracking and Motion. 정보산업공학협동과정 정우정. Contents. The Basics of Tracking Corner Finding Subpixel Corners Optical Flow Mean-Shift and Camshift Tracking Motion Templates Estimators The Condensation Algorithm. example. … /opencv/samples/c/ lkdemo.c (optical flow)
E N D
Tracking and Motion 정보산업공학협동과정 정우정
Contents • The Basics of Tracking • Corner Finding • Subpixel Corners • Optical Flow • Mean-Shift and Camshift Tracking • Motion Templates • Estimators • The Condensation Algorithm
example …/opencv/samples/c/ • lkdemo.c (optical flow) • camshiftdemo.c (mean-shift tracking of colored regions) • motempl.c (motion template) • kalman.c (Kalman filter)
The Basics of Tracking • Understand the motion of object:identification and modeling • IdentificationTo finding the object of interest from the video stream. • ModelingProviding us with noisy measurement of the object’s actual position.
Corner Finding (1/2) • Harris corner • Shi and Tomasi
Corner Finding (2/2) image: input image (single-channel) eigImage, tempImage: scratch by the algorithm corners: result points after the algorithm corner_count: maximum number of points quality_level: minimal eigenvalue (0<x<1) mask: usual image block_size: pixel use_harris: Harris or Shi-Tomasi k: weighting coefficient
Subpixel Corners (2/2) image: input corners: initial guesses for the corner location count: compute point win: size of window zero_zone: window that will not consider criteria: user-specified termination criterion
Optical Flow(Lucas-Kanade Method) (9/9) imgA: initial image imgB: final image pyrA, pyrB: buffers allocated to store the pyramid images featuresA: point for motion featuresB: new location point from featuresA count: number of points int the featureA winSize: window size level: depth of the stack of images status: 0/1 corresponding track_error: error value criteria: user-specified termination criterion flags: allow for some fine control
Mean-Shift and Camshift Tracking (1/3) Camshift: Continuously Adaptive Mean Shift Algorithm
Mean-Shift and Camshift Tracking (3/3) prob_image: density of probable locations window: kernel window criteria: user-specified termination criterion comp: converged search window location box: contain the newly resized box