INRIA Person Datat(法国国家信息与自动化研究所
行人数据库)
数据摘要：全国大学生英语四六级
This datat was collected as part of rearch work on detection of upright people in images and video. The rearch is described in detail in CVPR 2005 paper Histograms of Oriented Gradients for Human Detection and my PhD thesis. The datat is divided in two formats: (a) original images with corresponding annotation files, and (b) positive images in normalized 64x128 pixel format (as ud in the CVPR paper) with original negative images.
The data t contains images from veral different sources:
Images from GRAZ 01 datat, though annotation files are completely new.
坏账准备是什么科目Images from personal digital image collections taken over a long time period. Usually the original positive images were of very high resolution (approx. 2592x1944 pixels), so we have cropped the images to highlight persons. Many people are bystanders taken from the backgrounds of the input photos, so ideally there is no particular bias in their po.
Few of images are taken from the web using google images.
中文关键词：
直立，行人，检测，标注，规范化,
英文关键词：
upright,people,detection,annotation,normalized,
数据格式：
IMAGE
consider数据用途：
Rearch work on detection of upright people in images and video
数据详细介绍：
INRIA Person Datat
This datat was collected as part of rearch work on detection of upright people in images and video. The rearch is described in detail in CVPR 2005 paper Histograms of Oriented Gradients for Human Detection and my PhD thesis. The datat is divided in two formats: (a) original images with corresponding annotation files, and (b) positive images in normalized 64x128 pixel format (as ud in the CVPR paper) with original negative images. Contributions
The data t contains images from veral different sources:
crumbleImages from GRAZ 01 datat, though annotation files are completely new. Images from personal digital image collections taken over a long time period. Usually the original positive images were of very high resolution (approx. 2592x1944 pixels), so we have cropped the images to highlight persons. Many people are bystanders taken from the backgrounds of the input photos, so ideally there is no particular bias in their po.
Few of images are taken from the web using google images.imagine me without you
turbulence
Note
Only upright persons (with person height > 100) are marked in each image. Annotations may not be right; in particular at times portions of annotated bounding boxes may be outside or inside the object.
血腥打企鹅
Original Images
Folders 'Train' and 'Test' correspond, respectively, to original training and test images. Both folders have three sub folders: (a) 'pos' (positive training or test images), (b) 'neg' (negative training or test images), and (c) 'annotations' (annotation files for positive images in Pascal Challenge format).
Normalized Images
Folders 'train_64x128_H96' and 'test_64x128_H96' correspond to normalized datat as ud in above referenced paper. Both folders have two sub folders: (a) 'pos' (normalized positive training or test images centered on the person with their left-right reflections), (b) 'neg' (containing original negative training or test images). Note images in folder 'train/pos' are of 96x160 pixels (a margin of 16 pixels around each side), and images in folder 'test/pos' are of 70x134 pixels (a margin of 3 pixels around each side). This has been done to avoid boundary conditions (thus to avoid any particular bias in the classifier). In both folders, u the centered 64x128 pixels window for original detection task. Negative windows
asyncTo generate negative training windows from normalized images, a fixed t of 12180 windows (10 windows per negative image) are sampled randomly from 1218 negative training photos providing th
pplie initial negative training t. For each detector and parameter combination, a preliminary detector is trained and all negative training images are arched exhaustively (over a scale-space pyramid) for fal positives (`hard examples'). All examples with score greater than zero are considered hard examples. The method is then re-trained using this augmented t (initial 12180 + hard examples) to produce the final detector. The t of hard examples is subsampled if necessary, so that the descriptors of the final training t fit into 1.7 GB of RAM for SVM training. Starting scale in scale-space pyramid above is one and we keep adding one more level in the pyramid till floor(ImageWidth/Scale)>64 and floor(ImageHeight/Scale)>128. Scale ratio between two concutive levels in the pyramid is 1.2. Window stride (sampling distance between two concutive windows) at any scale is 8 pixels. If after fitting all windows at a scale level some margin remains at borders, we divide the margin by 2, take its floor and shift the whole window grid. For example, if image size at current level is (75,130), margin (with stride of 8 and window size of 64,128) left is (3,2). We shift all windows by (floor(MarginX/2), floor(MarginY/2)). New image width and height are calculated using the formulas: NewWidth=floor(OrigWidth/Scale)
and NewHeight=floor(OrigHeight/Scale). Here scale=1 implies the original image size.
destiny是什么意思
Also while testing negative images, to create negative windows, we u the same sampling structure.
You may download the whole data t from here (970MB). To avoid duplicating images, 'neg' image folder in 'train_64x128_H96' and 'test_64x128_H96' are referenced using symbolic links.
Disclaimer
THIS DATA SET IS PROVIDED "AS IS" AND WITHOUT ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, WITHOUT LIMITATION, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.
The images provided above may have certain copyright issues. We take no guarantees or responsibilities, whatsoever, arising out of any copyright issue. U at your own risk.
Support of European Union 6th framework project aceMedia is greatly acknowledged. For any questions, comments or other issues plea contact Navneet Dalal.
数据预览：
