[2005 CVPR] Histograms of Oriented Gradients for Human Detection

[2005 CVPR] Histograms of Oriented Gradients for Human Detection 用于人体检测的方向梯度直方图 Navneet Dalal，Bill Triggsigure 2. Some sample images from our new human detection database. The subjects are always upright, but with some partial occlusions and a wide range of variations in pose, appearance, clothing, illumination and background probabilities to be distinguished more easily. We will often strength and edge-presence based voting were tested, with use miss rate at 10FPPW as a reference point for results. the edge threshold chosen automatically to maximize detec This is arbitrary but no more so than, e.g. Area Under roc. tion performance(the values selected were somewhat vari- In a multiscale detector it corresponds to a raw error rate of able, in the region of 20-50 gray levels) bout 0. 8 false positives per 640 x 480 image tested. (The full Results. Fig. 3 shows the performance of the various detec- detector has an even lower false positive rate owing to non tors on the mit and inria data sets the hog-based de- maximum suppression). Our DET curves are usually quite tectors greatly outperform the wavelet, PCA-SIFT and Shape shallow so even very sinall improvements in miss rate are Context ones, giving near-perfect separation on the MIttest equivalent to large gains in FPPW at constant miss rate For set and at least an order of magnitude reduction in FPPw example, for our default detector at 1e-4 FPPW, every 1