1 / 14

Face Detection and Neural Networks

Face Detection and Neural Networks. Todd Wittman Math 8600: Image Analysis Prof. Jackie Shen December 2001. Face Detection. Problem: Given a color image, determine if the image contains a human face. That is, can you tell our governor from a toaster?. vs.

lotus
Télécharger la présentation

Face Detection and Neural Networks

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. Face DetectionandNeural Networks Todd Wittman Math 8600: Image Analysis Prof. Jackie Shen December 2001

  2. Face Detection Problem: Given a color image, determine if the image contains a human face. That is, can you tell our governor from a toaster? vs. Answer: The picture on the right contains a human face. I think. Applications: AI, tracking, automated security, video retrieval

  3. Overview of Face Detection Methods • Edge detection to recognize features and spatial relationships (Marsicoi, ‘97). • HSV-space segmentation and vector angular-based distance measure (Andoutsos, ‘99). • Chroma chart to detect skin tones and edge detection to identify eyes and mouth (Cai, ‘99). • Unsupervised Adaptive Skin Color Model, also called clustering (Bergasa, ‘99).

  4. Neural Network Goal: Given a set of inputs X and desired outputs T, determine the weights s.t. X generates T. Idea: Similar inputs will give similar outputs. X T Hidden Layer Training: Set weights to minimize . Levenberg-Marquad Algorithm (multi-dim steepest descent). Training is very expensive computationally. If there are x input nodes, t output nodes, and p hidden nodes, then # weights = (x+t)p.

  5. Face Detection NN Input: Color image. Output: P(w|x) = probability that image contains a face. (Only 1 output node.) Set 1 for face, 0 for no face. P=0 P=1 • 3 Possible Outputs • P > 0.5 FACE • P < 0.5 NOT FACE • P = 0.5 DON’T KNOW

  6. 1st Attempt: Interpolated Image Input X: The pixel values of the image at N selected grid points. Original Interpolated Output P=1 Since each pixel has three values (RGB), our input vector X will have length 3N. I tried a small case: N=25. The network took over an hour to train for the training set on the next slide.

  7. Training Set

  8. Results P values for 20 images in training set. P=0.5 for all training images. The interpolated images can’t be interpreted.

  9. 2nd Attempt: RGB Histograms Input X: The 3 histograms of the RGB values, appended as 1 vector. Each histogram has N=20 bins. So size of input vector is 3N=60. Idea: Neural network will pick out the frequency of flesh tones.

  10. Results After 100 iterations (1 hour, 1241 weights), the Levenberg-Marquad algorithm was able to correctly classify all 20 training images. 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 But on a test set of 13 images, got 7 correct (53.8%).

  11. 3rd Attempt: YES Histograms RGB histograms were too similar. Y = 0.253R + 0.684G + 0.063B E = 0.5R - 0.5G S = 0.25R + 0.25G - 0.5B RGB YES Input X: 3 YES histograms appended as one vector.

  12. Results After training for 100 iterations, 3 images in training set were mis-classified. 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 0 - 0.1 But on test set, correctly identified 13 out of 13 images (100%).

  13. Th-th-that’s All, Folks! You can try my Matlab code: www.math.umn.edu/~wittman/faces/main.html

  14. Input Layer Output Layer Hidden Layer T X

More Related