Showing papers by "Alexander C. Berg published in 2009"

Attribute and simile classifiers for face verification

Abstract: We present two novel methods for face verification. Our first method - “attribute” classifiers - uses binary classifiers trained to recognize the presence or absence of describable aspects of visual appearance (e.g., gender, race, and age). Our second method - “simile” classifiers - removes the manual labeling required for attribute classification and instead learns the similarity of faces, or regions of faces, to specific reference people. Neither method requires costly, often brittle, alignment between image pairs; yet, both methods produce compact visual descriptions, and work on real-world images. Furthermore, both the attribute and simile classifiers improve on the current state-of-the-art for the LFW data set, reducing the error rates compared to the current best by 23.92% and 26.34%, respectively, and 31.68% when combined. For further testing across pose, illumination, and expression, we introduce a new data set - termed PubFig - of real-world images of public figures (celebrities and politicians) acquired from the internet. This data set is both larger (60,000 images) and deeper (300 images per individual) than existing data sets of its kind. Finally, we present an evaluation of human performance.

Max-margin additive classifiers for detection

Abstract: We present methods for training high quality object detectors very quickly The core contribution is a pair of fast training algorithms for piece-wise linear classifiers, which can approximate arbitrary additive models The classifiers are trained in a max-margin framework and significantly outperform linear classifiers on a variety of vision datasets We report experimental results quantifying training time and accuracy on image classification tasks and pedestrian detection, including detection results better than the best previous on the INRIA dataset with faster training

Finding iconic images

Abstract: We demonstrate that is it possible to automatically find representative example images of a specified object category. These canonical examples are perhaps the kind of images that one would show a child to teach them what, for example a horse is - images with a large object clearly separated from the background. Given a large collection of images returned by a web search for an object category, our approach proceeds without any user supplied training data for the category. First images are ranked according to a category independent composition model that predicts whether they contain a large clearly depicted object, and outputs an estimated location of that object. Then local features calculated on the proposed object regions are used to eliminate images not distinctive to the category and to cluster images by similarity of object appearance. We present results and a user evaluation on a variety of object categories, demonstrating the effectiveness of the approach.

Methods, systems, and media for automatically classifying face images

Abstract: Methods, systems, and media for automatically classifying face images are provided. In some embodiments, features of the face image to be classified for an attribute are selected, wherein each of the features corresponds to a different region of the face image and specifies one or more of a type of pixel data to be evaluated for the region, a normalization to be applied for the region, and an aggregation to be applied for the region. The face image is classified with respect to the attribute based on the features of the image, and the attribute and a confidence value are assigned to the face image based on the classifying. A query is received from a user, and the attribute is identified as corresponding to the query. The face image is determined as corresponding to the attribute, and the face image is identified to the user as corresponding to the query.