18类物品数据库(手动标注)(18 Object-Class Databases (manually annotated))_图像处理_科研数据集.pdf

18类物品数据库(手动标注)(18 Object-Class Databases (manually annotated))_图像处理_科研数据集.pdf

  1. 1、本文档共7页,可阅读全部内容。
  2. 2、原创力文档(book118)网站文档一经付费(服务费),不意味着购买了该文档的版权,仅供个人/单位学习、研究之用,不得用于商业用途,未经授权,严禁复制、发行、汇编、翻译或者网络传播等,侵权必究。
  3. 3、本站所有内容均由合作方或网友上传,本站不对文档的完整性、权威性及其观点立场正确性做任何保证或承诺!文档内容仅供研究参考,付费前请自行鉴别。如您付费,意味着您自己接受本站规则且自行承担风险,本站不退款、不进行额外附加服务;查看《如何避免下载的几个坑》。如果您已付费下载过本站文档,您可以点击 这里二次下载
  4. 4、如文档侵犯商业秘密、侵犯著作权、侵犯人身权等,请点击“版权申诉”(推荐),也可以打举报电话:400-050-0827(电话支持时间:9:00-18:30)。
查看更多
18类物品数据库(手动标注)(18 Object-Class Databases (manually annotated))_图像处理_科研数据集

18 类物品数据库(手动标注)(18 Object-Class Databases (manually annotated)) 数据介绍: airplane, beaver, bikes, boat, camel, car, dolphin, elephant, giraffe, guitar, horse, kangaroo, motorbikes, penguin, shark, tiger, wristwatch, zebra 关键词: 交通工具,动物,吉他,手表,手动标注, vehicle,animal,guitar,wristwatch,manually annotated, 数据格式: IMAGE 数据详细介绍: 18 Object-Class Databases (manually annotated) Florian Schroff, Antonio Criminisi and Andrew Zisserman Please note the license for this data. Image Preview There is a preview showing the images arranged by their manual annotation: good: links to the images labelled as good. ok: links to the images labelled as ok junk : links to the images not labelled good or ok abstr.: links to the images labelled as abstract good or abstract ok Zip-Files Each zip-file contains all images as well as the associated text-files. Furthermore, there are two subdirectories ranking and stats. ranking: the first file contains the images ranked by our naive Bayes ranker; the second file (suffix _rd.txt) contains this same ranking, but contains o nly one image per webpage; both files contain filtered images only (see section 2.1 in the paper). stats : contains a matlab file with a matrix (imagestats) that represents some of the parsed content of the textfiles as specified in documentlist.txt; the corresponding entries for the metadata fields (row 1:8, as specified by the filename) are 1 if the query-word occurs and 0 otherwise (see paper); Row 9,10 contain a coding for the image-extension and mime-type and the remaining rows contain the annotation and the rank at which Google imagesearch returned each image; the last row describes a binary coding of the service (see paper, table 1). 18 Object-Class Databases (manually annotated) (airplane, beaver, bikes, boat, camel, car, dolphin, elephant, giraffe, guitar, horse, kangaroo, motorbikes, penguin, shark, tiger, wristwatch, zebra)  airplane.zi

文档评论(0)

yan698698 + 关注
实名认证
内容提供者

该用户很懒,什么也没介绍

1亿VIP精品文档

相关文档