21
Presentation 4 Ruben Villegas 06/12/2012

Presentation 4 Ruben Villegas 06/12/2012

  • Upload
    bela

  • View
    37

  • Download
    0

Embed Size (px)

DESCRIPTION

Presentation 4 Ruben Villegas 06/12/2012. Approximating crowd density taking advantage of occlusion. Papers read. - PowerPoint PPT Presentation

Citation preview

Page 1: Presentation 4 Ruben Villegas 06/12/2012

Presentation 4Ruben Villegas06/12/2012

Page 2: Presentation 4 Ruben Villegas 06/12/2012

Approximating crowd density taking advantage of occlusion

Page 3: Presentation 4 Ruben Villegas 06/12/2012

Papers readDensity-aware person detection

and tracking in crowds by Mikel Rodriguez, Ivan Laptev, Josef Sivic, Jean-Yves Audibert, Ecole Normale Superieure, INRIA Imagine, LIGM, UniversiteParis-Est

Part-based Multiple-Person Tracking with Partial Occlusion Handling

Page 4: Presentation 4 Ruben Villegas 06/12/2012

Getting the Occlusion Map from the Score Map

Page 5: Presentation 4 Ruben Villegas 06/12/2012

Score map

Page 6: Presentation 4 Ruben Villegas 06/12/2012

Clustering of different score areas using k-means, k = 4

Page 7: Presentation 4 Ruben Villegas 06/12/2012

Score map valuesThe intense red in the score map

represents high detection scores returned by the part-base model in the algorithm

The blue represents low detection scores.

The green and yellow are assumed to be only parts of humans but not the entire body, since the scores sit in the middle

Page 8: Presentation 4 Ruben Villegas 06/12/2012

First try to get the occlusion mapscore.*inv_scoreThe assumption was that this will multiply high

values with high values inverted and same for low values, therefore the middle values will stay.

Since inv_score is (score)^(-1) before being converted to range between 0 nad 1, if the max value in score was 5, inv_score = 0.2, and in score converted between 0 and 1, 5 will be 1. Therefore, 5 will map to 1 and the middle values will be ie. 0.3*0.5 = 0.18 and 1*0.25 = 0.25 we do not accomplish our goal of making middle values higher.

Page 9: Presentation 4 Ruben Villegas 06/12/2012

Process to get middle scores

Get the max and min values from score.

Go through every element in score and subtract from max and min separately.

Compare results and pick the lowest one.

Save all those results in a new matrix.

Page 10: Presentation 4 Ruben Villegas 06/12/2012

Middle values

Page 11: Presentation 4 Ruben Villegas 06/12/2012

Getting approximate occlusion mapTo get the occlusion map, we want to

make sure the partially detected people are pronounced.

Adding the score map and the middle values map will give us an approximate occlusion map

To check the result it is used as a mask on the original picture keeping the values that are higher than the average of the max and min values in the occlusion map.

Page 12: Presentation 4 Ruben Villegas 06/12/2012

Occlusion map

Page 13: Presentation 4 Ruben Villegas 06/12/2012

Occlusion map masked on image

Page 14: Presentation 4 Ruben Villegas 06/12/2012

What’s wrong with the occlusion map?For every true detection(high

scores), there are many middle values around them which refer to the same people, and they shouldn’t be considered in the occlusion map.

To solve this we will incorporate the scores from each part separately in the occlusion map.

Page 15: Presentation 4 Ruben Villegas 06/12/2012

Incorporating the partsEvery pixel in the score map

represents the sum of the score of all the parts detection scores.

Incorporating the score for each part directly into the occlusion map will give us a better occlusion map by emphasizing each part in it.

Page 16: Presentation 4 Ruben Villegas 06/12/2012

Parts score maps

Page 17: Presentation 4 Ruben Villegas 06/12/2012

All parts score map added up

Page 18: Presentation 4 Ruben Villegas 06/12/2012

All parts score map added up

Score map from the entire Model detection

Page 19: Presentation 4 Ruben Villegas 06/12/2012

Problems with Pedro’s code

Page 20: Presentation 4 Ruben Villegas 06/12/2012
Page 21: Presentation 4 Ruben Villegas 06/12/2012