Citation Context

...d-alone sections were specifically devoted to local matching methods [43], [54]. Careful comparative studies of different options in a holistic recognition system have been reported in the literature =-=[38]-=-. We believe that a similar comparative study on the options at each step in the local matching process will benefit face recognition through localized matching. Although several stand-alone local mat...

Citation Context

...ional, and therefore, from (i) and (ii), almost constant in W. Another interesting algorithm is the so-called plane subtraction or illumination gradient compensation algorithm (Sung and Poggio, 1998; =-=Ruiz-del-Solar and Navarrete, 2005-=-), which consists on calculating a best-fit brightness plane to the image under analysis, and then subtracting this plane to the image. This method allows compensating heavy shadows caused by extreme ...

Citation Context

...ion from d to d ′ ,whered ′ <d,sothatinR d′ the within-class scatter matrix is not singular. The FisherFace [1], [19], the most discriminant features [32], QR-decomposition [42], the subspace methods =-=[31]-=-, [44] and recursive LDA [39] have been developed. These approaches are straightforward, but some of them suffer from two limitations. First, the FisherFace method might fail [45]. Selections of featu...

Citation Context

...ocess. To adequately service such a large website, an auto-tagging system must not only exhibit high accuracy, but high speed, low memory consumption, and scalability. Ruizdel-Solar and Navarrette in =-=[10]-=- provide an excellent review of face recognition algorithms, but only evaluate them for accuracy. Our contribution is utilizing a new source of real-world data to evaluate several face recognition alg...

Citation Context

...h linear classifiers.sThese methods were selected by considering their fulfillment of the requirementssmentioned above, and their performance in former comparative studies of facesrecognition methods =-=[6]-=-[7][13][15].sThe comparative study is carried out using the LFW database [14] (the restrictedssetting was considered, see below for details). Aspects such as scale, pose, lighting,sfocus, resolution, ...

Citation Context

... these ideas in mind, we have include in our Face and Hand Analysis module the following components: face detection (using boosted classifiers) [2][3], face recognition (using eigenspace projections) =-=[4]-=-[5], people tracking (using face information and Kalman Filters) [7][8], gender classification using facial information [1], age classification using facial information, skin detection [6], hand detec...

Citation Context

...ity images [6,7,8,9,10]: Holistic matching methods use the whole face region as the raw input to a recognition system. One of the most widely used representations of the face region is eigen pictures =-=[11,12,13,14,15,16]-=-, which are based on Principal Component Analysis (PCA). Using PCA, many face recognition techniques have been developed: eigenfaces, which use a nearest neighbor classifier; feature-line-based method...

Citation Context

...he “Eigenfaces” and the “Fisherfaces” can be considered the two main subspace decomposition work that have been improved upon by other researchers. Comparative studies by Ruiz-del-Solar and Navarrete =-=[3]-=-, Delac et al. [4] and G. Shakhnarovich and Moghaddam [11] to name a few, might direct interested readers to the available subspace techniques. The underlying ideology among the subspace techniques is...

Citation Context

... the recognition method needs to be robust to noise, pose, illumination and expression (PIE) variations, in order to allow it to be novel and to compete with the other modern face recognition methods =-=[12]-=-, [13] which can’t handle all of these variations simultaneously [14]. As pose variations are rigid body transformations, the relative distances and angles between the salient points are unaffected by...

Citation Context

...r age of the user. In this context our group has a successful record in developing face analysis algorithms. So far we have developed automated tools for face detection [17][22], face recognition [18]=-=[10]-=-, people tracking [32][35], gender classification [5], and skin detection [9][28][31]. We have also integrated these tools in robot control systems, for instance AIBO ERS7 robots [35]. Demos of our fa...