收藏 分享(赏)

iola and Micha.ppt

上传人:hwpkd79526 文档编号:7322028 上传时间:2019-05-14 格式:PPT 页数:18 大小:273KB
下载 相关 举报
iola and Micha.ppt_第1页
第1页 / 共18页
iola and Micha.ppt_第2页
第2页 / 共18页
iola and Micha.ppt_第3页
第3页 / 共18页
iola and Micha.ppt_第4页
第4页 / 共18页
iola and Micha.ppt_第5页
第5页 / 共18页
点击查看更多>>
资源描述

1、Robust Real-time Object Detection by Paul Viola and Michael Jones ICCV 2001 Workshop on Statistical and Computation Theories of Vision,Presentation by Gyozo Gidofalvi Computer Science and Engineering Department University of California, San Diego gyozocs.ucsd.edu October 25, 2001,Outline,Object dete

2、ction task Definition and rapid evaluation of simple features for object detection Method for classification and feature selection, a variant of AdaBoost Speed-up through the Attentional Cascade Experiments and Results Conclusions,Object detection task,Object detection framework: Given a set of imag

3、es find regions in these images which contain instances of a certain kind of object. Task: Develop an algorithm to learn an fast and accurate method for object detection.To capture ad-hoc domain knowledge classifiers for images do not operate on raw grayscale pixel values but rather on values obtain

4、ed from applying simple filters to the pixels.,Definition of simple features for object detection,3 rectangular features types:two-rectangle feature type (horizontal/vertical)three-rectangle feature typefour-rectangle feature type,Using a 24x24 pixel base detection window, with all the possible comb

5、ination of horizontal and vertical location and scale of these feature types the full set of features has 49,396 features. The motivation behind using rectangular features, as opposed to more expressive steerable filters is due to their extreme computational efficiency.,Integral image,Def: The integ

6、ral image at location (x,y), is the sum of the pixel values above and to the left of (x,y), inclusive. Using the following two recurrences, where i(x,y) is the pixel value of original image at the given location and s(x,y) is the cumulative column sum, we can calculate the integral image representat

7、ion of the image in a single pass.,(x,y),s(x,y) = s(x,y-1) + i(x,y) ii(x,y) = ii(x-1,y) + s(x,y),(0,0),x,y,Rapid evaluation of rectangular features,Using the integral image representation one can compute the value of any rectangular sum in constant time. For example the integral sum inside rectangle

8、 D we can compute as: ii(4) + ii(1) ii(2) ii(3),As a result two-, three-, and four-rectangular features can be computed with 6, 8 and 9 array references respectively.,Challenges for learning a classification function,Given a feature set and labeled training set of images one can apply number of mach

9、ine learning techniques. Recall however, that there is 45,396 features associated with each image sub-window, hence the computation of all features is computationally prohibitive. Hypothesis: A combination of only a small number of these features can yield an effective classifier. Challenge: Find th

10、ese discriminant features.,A variant of AdaBoost for aggressive feature selection,Performance of 200 feature face detector,The ROC curve of the constructed classifies indicates that a reasonable detection rate of 0.95 can be achieved while maintaining an extremely low false positive rate of approxim

11、ately 10-4.,First features selected by AdaBoost are meaningful and have high discriminative powerBy varying the threshold of the final classifier one can construct a two-feature classifier which has a detection rate of 1 and a false positive rate of 0.4.,Speed-up through the Attentional Cascade,Simp

12、le, boosted classifiers can reject many of negative sub-windows while detecting all positive instances.Series of such simple classifiers can achieve good detection performance while eliminating the need for further processing of negative sub-windows.,Processing in / training of the Attentional Casca

13、de,Processing: is essentially identical to the processing performed by a degenerate decision tree, namely only a positive result from a previous classifier triggers the evaluation of the subsequent classifier. Training: is also much like the training of a decision tree, namely subsequent classifiers

14、 are trained only on examples which pass through all the previous classifiers. Hence the task faced by classifiers further down the cascade is more difficult. To achieve efficient cascade for a given false positive rate F and detection rate D we would like to minimize the expected number of features

15、 evaluated N:,Since this optimization is extremely difficult the usual framework is to choose a minimal acceptable false positive and detection rate per layer.,Algorithm for training a cascade of classifiers,Experiments (dataset for training),4916 positive training example were hand picked aligned,

16、normalized, and scaled to a base resolution of 24x24 10,000 negative examples were selected by randomly picking sub-windows from 9500 images which did not contain faces,Experiments cont. (structure of the detector cascade),The final detector had 32 layers and 4297 features total,Speed of the detecto

17、r total number of features evaluatedOn the MIT-CMU test set the average number of features evaluated is 8 (out of 4297).The processing time of a 384 by 288 pixel image on a conventional personal computer about .067 seconds.Processing time should linearly scale with image size, hence processing of a

18、3.1 mega pixel images taken from a digital camera should approximately take 2 seconds.,Operation of the face detector,Since training examples were normalized, image sub-windows needed to be normalized also. This normalization of images can be efficiently done using two integral images (regular / squ

19、ared). Detection at multiple scales is achieved by scaling the detector itself. The amount of shift between subsequent sub-windows is determined by some constant number of pixels and the current scale. Multiple detections of a face, due to the insensitivity to small changes in the image of the final

20、 detector were, were combined based on overlapping bounding region.,Results,Testing of the final face detector was performed using the MIT+CMU frontal face test which consists of:130 images505 labeled frontal faces Results in the table compare the performance of the detector to best face detectors k

21、nown.,Rowley at al.: use a combination of 1wo neural networks (simple network for prescreening larger regions, complex network for detection of faces). Schneiderman at al.: use a set of models to capture the variation in facial appearance; each model describes the statistical behavior of a group of

22、wavelet coefficients.,Results cont.,Conclusion,The paper presents general object detection method which is illustrated on the face detection task. Using the integral image representation and simple rectangular features eliminate the need of expensive calculation of multi-scale image pyramid. Simple

23、modification to AdaBoost gives a general technique for efficient feature selection. A general technique for constructing a cascade of homogeneous classifiers is presented, which can reject most of the negative examples at early stages of processing thereby significantly reducing computation time. A face detector using these techniques is presented which is comparable in classification performance to, and orders of magnitude faster than the best detectors know today.,

展开阅读全文
相关资源
猜你喜欢
相关搜索

当前位置:首页 > 企业管理 > 管理学资料

本站链接:文库   一言   我酷   合作


客服QQ:2549714901微博号:道客多多官方知乎号:道客多多

经营许可证编号: 粤ICP备2021046453号世界地图

道客多多©版权所有2020-2025营业执照举报