1 / 39

Tracking with Local Spatio -Temporal Motion Patterns in Extremely Crowded Scenes

Tracking with Local Spatio -Temporal Motion Patterns in Extremely Crowded Scenes. Present by 陳群元. outline. Introduction Previous work Predicting motion patterns Spatio -temporal transition distribution Discerning pedestrians Experimental results conclusion. introduction.

lynsey
Download Presentation

Tracking with Local Spatio -Temporal Motion Patterns in Extremely Crowded Scenes

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript


  1. Tracking with Local Spatio-Temporal Motion Patternsin Extremely Crowded Scenes Present by 陳群元

  2. outline • Introduction • Previous work • Predicting motion patterns • Spatio-temporal transition distribution • Discerning pedestrians • Experimental results • conclusion

  3. introduction • Tracking individuals in extremely crowded scenes is a challenging task, • we predict the local spatio-temporal motion patterns that describe the pedestrian movement at each space-time location in the video. • we robustly model the individual’s unique motion and appearance to discern them from surrounding pedestrians.

  4. Previous work • Previous work track features and associate similar trajectories to detect individual moving entities within crowded scenes. • We encode many possible motions in the HMM, and derive a full distribution of the motion at each spatio-temporal location in the video.

  5. outline • Introduction • Previous work • Predicting motion patterns • Spatio-temporal transition distribution • Discerning pedestrians • Experimental results • conclusion

  6. Markov Model 0.6 • An example : a 3-state Markov Chain λ • State 1 generates symbol A only, State 2 generates symbol B only, and State 3 generates symbol C only • Given a sequence of observed symbols O={CABBCABC}, the only one corresponding state sequence is {S3S1S2S2S3S1S2S3}, and the corresponding probability isP(O|λ)=P(q0=S3) P(S1|S3)P(S2|S1)P(S2|S2)P(S3|S2)P(S1|S3)P(S2|S1)P(S3|S2) =0.10.30.30.70.20.30.30.2=0.00002268 s1 A 0.3 0.3 0.3 0.1 0.2 0.7 0.7 s2 s3 0.2 C B

  7. 0.6 s1 {A:.3,B:.2,C:.5} 0.3 0.3 0.3 0.1 0.2 0.7 0.7 s2 s3 0.2 {A:.7,B:.1,C:.2} {A:.3,B:.6,C:.1} Hidden Markov Model • An example : a 3-state discrete HMM λ • Given a sequence of observations O={ABC}, there are 27 possible corresponding state sequences, and therefore the corresponding probability is

  8. Predicting Motion Patterns

  9. Spatio-temporal gradient • f(Pos) = (f(Pos+1) -f(Pos) + f(Pos) -f(Pos-1))/2 = f(Pos+1)-f(Pos-1)/2; • For each pixel i in cuboid • I is intensity

  10. spatio-temporal motion pattern • the local spatio-temporal motion pattern • represented by a 3D Gaussian of spatio-temporal gradients

  11. Training HMM • The hidden states of the HMM are represented by a set of motion patterns • The probabilityof an observed motion pattern given a hidden state s is

  12. Kullback–Leiblerdivergence • Kullback–Leiblerdivergence is a non-symmetric measure of the difference between two probability distributions P and Q.

  13. predictive distribution • After training a collection of HMMs on a video of typicalcrowd motion, we predict the motion pattern at eachspace-time location that contains the tracked subject. • where S is the set of hidden states, w(s) is defined by

  14. Vector of scaled message Reference :A Tutorial On Hidden Markov Models andSelected Applications in Speech Recognition.

  15. predicted localspatio-temporal motion pattern • a weighted sum of the 3D Gaussian distributionsassociated with the HMM’s hidden states

  16. The centroid we are interested in is a multivariate normal density that minimizes the total distortions. Formally, a centroid c is defined as, Reference: On Divergence Based Clustering of Normal Distributions and Its Application to HMM Adaptation

  17. Predicted motion pattern • where and are the mean and covariance of the hidden state s, respectively.

  18. outline • Introduction • Previous work • Predicting motion patterns • Spatio-temporal transition distribution • Discerning pedestrians • Experimental results • conclusion

  19. Bayesian probabilities

  20. we use the gradient information to estimate the opticalflow within each specific sub-volume and track the target ina Bayesian framework. • Bayesian tracking can be formulated as maximizingthe posterior distribution of the state xt of the target at timet given available measurements z1:t= {zi; i = 1 : : : t} by • zt is the image at time t, p (xt|xt-1) is the transition distribution, and p (zt|xt) is the likelihood. • state vector x t as the width, height, and 2D location of the target within the image.

  21. we focus on the target’smovement between frames and use a 2nd-degree autoregressivemodel for the transition distribution of thetarget’s width and height. • Ideally, the state transition distribution p (xt|xt-1) directly reflects the two-dimensional motion of the target between frames t -1 and t. • where is the 2D optical flow vector, and is the covariance matrix.

  22. optical flow • Assuming the movement to be small, the image constraint at I(x,y,t) with Taylor series can be developed to get • H.O.T

  23. The predicted motion pattern is defined by a mean gradient vector and a covariance matrix • The motion information encoded in the spatio-temporal gradients can be expressed in the form of the structure tensor matrix • The optical flow can then be estimated from the structure tensor by solving • where w = [u; v; z]T is the 3D optical flow

  24. Covariance matrix

  25. outline • Introduction • Previous work • Predicting motion patterns • Spatio-temporal transition distribution • Discerning pedestrians • Experimental results • conclusion

  26. Typical models of the likelihood distribution p (z t |x t ) • where is the variance, is a distance measure, and Z is a normalization term. • difference between aregion R (defined by state x t ) of the observed image z t andthe template. • We assume pedestrians exhibit consistency in their appearance and their motion, and model them in a joint likelihood by • where pA and pM are the appearance and motion likelihoods

  27. Update motion template • After tracking in frame t, we update each pixel i in the motion template by • where is the motion template at time t, • Is the region of spatio-temporal gradient defined by the tracking result (i.e., the expected value of the posterior) • is the learning rate.

  28. update this error measurement • The error at pixel i and time t becomes • ti and ri are the normalized gradient vectors of the motion template and the tracking result at time t • To reduce the contributions of frequently changing pixels to the computation of the motion likelihood, we weigh each pixel in the likelihood’s distance measure. • where Z is a normalization term such that

  29. distance measure • The distance measure of the motion likelihood distribution becomes

  30. outline • Introduction • Previous work • Predicting motion patterns • Spatio-temporal transition distribution • Discerning pedestrians • Experimental results • conclusion

  31. The training video for the concourse scene contains 300 frames (about 10 seconds of video), • the video for ticket gate scene contains 350 frames. • We set the cuboid size to 10*10*10 for both scenes. • The learning rate , appearance variance , and motion variance are 0.05.

  32. outline • Introduction • Previous work • Predicting motion patterns • Spatio-temporal transition distribution • Discerning pedestrians • Experimental results • conclusion

  33. Conclusion • In this paper, we derived a novel probabilistic method that exploits the inherent spatially and temporally varying structured pattern of a crowd’s motion to track individuals in extremely crowded scenes.

  34. The end • Thank you

More Related