SVHN街景门牌号数据集
收藏资源简介:
SVHN is a real-world image dataset for developing machine learning and object recognition algorithms with minimal requirement on data preprocessing and formatting. It can be seen as similar in flavor toMNIST(e.g., the images are of small cropped digits), but incorporates an order of magnitude more labeled data (over 600,000 digit images) and comes from a significantly harder, unsolved, real world problem (recognizing digits and numbers in natural scene images). SVHN is obtained from house numbers in Google Street View images.ReferencePlease cite the following reference in papers using this dataset:Yuval Netzer, Tao Wang, Adam Coates, Alessandro Bissacco, Bo Wu, Andrew Y. NgReading Digits in Natural Images with Unsupervised Feature LearningNIPS Workshop on Deep Learning and Unsupervised Feature Learning 2011. (PDF)Please usehttp://ufldl.stanford.edu/housenumbersas the URL for this site when necessaryFor questions regarding the dataset, please contactstreetviewhousenumbers@gmail.com
SVHN是一款用于开发机器学习与目标识别算法的真实世界图像数据集,对数据预处理与格式调整的要求极低。该数据集在特性上可类比MNIST,例如其图像均为经裁剪的小型数字图像,但标注数据量高出一个数量级以上,总计超过60万张数字图像,且源自一个难度显著更高、尚未完全解决的真实世界问题:自然场景图像中的数字与号码识别。SVHN数据集的图像素材来源于谷歌街景(Google Street View)中的房屋门牌号。 引用须知:若使用该数据集开展论文研究,请引用以下文献:Yuval Netzer、Tao Wang、Adam Coates、Alessandro Bissacco、Bo Wu、Andrew Y. Ng,《基于无监督特征学习的自然图像数字识别》,2011年神经信息处理系统研讨会(NIPS)深度学习与无监督特征学习分论坛(PDF)。必要时,请以http://ufldl.stanford.edu/housenumbers作为该数据集官方网站的访问地址。若有关于该数据集的相关疑问,请发送邮件至streetviewhousenumbers@gmail.com




