《人脸识别的简单算法》由会员分享，可在线阅读，更多相关《人脸识别的简单算法（17页珍藏版）》请在金锄头文库上搜索。
1、Rowley-Baluja-Kanade Face DetectorAuthor: Scott SannerContents.Introduction Algorithm. Data Preparation Training Image Scanning Testing Conclusion References SoftwareIntroductionThe goal of this project is to implement and analyze theRowley-Baluja-Kanade neural net face detector as described in 2 al
2、ong with some enhancements for training and recognition proposed by Sung and Poggio as described in 3. The basic goal underlying both approaches is to train a neural network or other recognition system on a labelled database of face and non-face images. This face classifier can then be usedtoscanove
3、ran imageresolutionpyramidtodeterminethelocations and scaling of any faces (if present) and return them to the user.Overall, the task of face recognition can be extremely difficult given the wide variety of faces to match, the presence of facial hair, variations in lighting and shadowing, and the po
4、ssibility of angular, scaling, and dimensional variances. Consequently an ideal face detector should attempt to mitigate all of these problems while achieving a high detection rate and minimizing the number of false positives. As we will see in the latter requirement, there is a tradeoff between the
5、 positive detection rate and the false positive rate and the balance between the two will need to be evaluated by the individual user and application domain.Algorithm OverviewTo achieve the above goals for face detection, we use a general algorithm that is a straightforward application of data prepa
6、ration, training, and image scanning. This algorithm is outlined below:Normalize Training Data:- For each face and non-face image:- Subtract out an approximation of the shading plane to correct for single light source effects- Rescale histogram so that every image has the same same gray level range-
7、 Aggregate data into labeled data setsTrain Neural Net:- Until the Neural Net reaches convergence (or a decrease in performance on the validation set):- Perform gradient descent error backpropagation on on the neural net for the batch of all training dataApply Face Detector to Image:- Build a resolu
8、tion pyramid of the image by successively successively decreasing the image resolution at each level of the pyramid, stopping at some default minimum resolution- For each level of the pyramid- Scan over the image, applying the trained neural net face detector to each rectangle within the image- If a
9、 positive face classification is found for a rectangle, scale this rectangle to the size appropriate for the original image and add it to the face bounding-box set- Return the rectangles in the face bounding-box setData PreparationIn performing face detection with a neural net, a few face-specific a
10、nd non-face-specific issues arise.In the realm of face specific issues, we do not want the background to become involved in face matching. Consequently, if person A is in two different settings wewant to ensure that we perform as well as possible in detecting person As face despite the background va
11、riation. If we were only to look at potential candidate rectangles for a face then we would receive interference from the corners which are more likely to consist of background than face pixels. Neural nets are especially susceptible to such errors since any consistencies between data in the trainin
12、g set (no matter how plausible a predictor of face-hood in real life) will likely be detected and expl oit ed. Thus, as 3 sugges ts, it is a good idea to maskanovalwithinthefacerectangletoprunethepixelsusedintraining in neural net. For true face images, this usually guarantees that only pixels from
13、the face are used as input to the neural net. For our implementation, we use the oval mask which can be seen in figure 3. The bounding rectangle for this mask is 18 x 27 pixels.Another face specific issue is that of pose or glasses. We want to recognize a face invariant of whether a person is smilin
14、g, sad, wearing glasses,ornotwearingglasses.Consequentlyitisimportanttoconstruct a set of training data which covers a broad range of human emotions, poses, and glasses/non-glasses wearing faces. This ensures the greatest generalization when applying the face detector to faces which have not been se
15、en before. For our dataset, we use 30 faces and their left-right flippedversionswithavarietyofemotionsandposesascontainedinthe Yale Face Dat abase 1. It would be adva nt ageous to have more faces and posesthanthisbutthetimelimitsofthisprojectconstrainedtheamount of time that could be devoted to phot
16、oediting (since the Yale Face Database is not in a directly usable format).One non-face specific issue is that of lighting direction. Neural nets are especially susceptible to pixel magnitude values and the differences between images illuminated from the left or right may be enough to make them appear as two different classifications from the perspective of the neural net. Consequently, there has
