collection curated by carandraug https://academictorrents.com/collection/vgg-oxford @article{, title= {Learning a Part-Level Motion Prior for Articulated Objects}, journal= {}, author= {Ruining Li and Chuanxia Zheng and Christian Rupprecht and Andrea Vedaldi}, year= {}, url= {}, abstract= {}, keywords= {}, terms= {}, license= {}, superseded= {} } https://academictorrents.com/download/2e955e41f40147603641573b7e839efae9af9a7f We have created a 37 category pet dataset with roughly 200 images for each class. The images have a large variations in scale, pose and lighting. All images have an associated ground truth annotation of breed, head ROI, and pixel level trimap segmentation. https://academictorrents.com/download/b18bbd9ba03d50b0f7f479acc9f4228a408cecc1 This is a synthetically generated dataset, in which word instances are placed in natural scene images, while taking into account the scene layout. The dataset consists of *800 thousand* images with approximately *8 million* synthetic word instances. Each text instance is annotated with its text-string, word-level and character-level bounding-boxes. https://academictorrents.com/download/2dba9518166cbd141534cbf381aa3e99a087e83c The exact data used to train our deep convolutional neural networks (see our [research page]()) is included in this torrent. This is synthetically generated dataset which we found sufficient for training text recognition on real-world images ![Synthetic Data Engine processt]() This dataset consists of *9 million images* covering *90k English words*, and includes the training, validation and test splits used in our work. https://academictorrents.com/download/3d0b4f09080703d2a9c6be50715b46389fdb3af1