- Referenced in 167 articles
- images per class. There are 50000 training images and 10000 test images. The dataset ... exactly 1000 randomly-selected images from each class. The training batches contain the remaining images ... random order, but some training batches may contain more images from one class than another ... images each. There are 500 training images and 100 testing images per class...
- Referenced in 474 articles
- version of the network trained on more than a million images from the ImageNet database...
- Referenced in 108 articles
- networks for biomedical image segmentation. There is large consent that successful training of deep networks ... network can be trained end-to-end from very few images and outperforms the prior ... Using the same network trained on transmitted light microscopy images (phase contrast ... network is fast. Segmentation of a 512x512 image takes less than a second...
- Referenced in 49 articles
- categories, with 7,000 images per category. The training set has 60,000 images ... shares the same image size, data format and the structure of training and testing splits...
- Referenced in 17 articles
- nearest neighbor models for image auto-annotation. Image auto-annotation is an important open problem ... TagProp, a discriminatively trained nearest neighbor model. Tags of test images are predicted using ... nearest-neighbor model to exploit labeled training images. Neighbor weights are based on neighbor rank ... training set. In this manner, we can optimally combine a collection of image similarity metrics...
- Referenced in 45 articles
- image denoising. Specifically, residual learning and batch normalization are utilized to speed up the training ... existing discriminative denoising models which usually train a specific model for additive white Gaussian noise ... latent clean image in the hidden layers. This property motivates us to train a single...
- Referenced in 289 articles
- digits, available from this page, has a training set of 60,000 examples ... normalized and centered in a fixed-size image. It is a good database for people...
- Referenced in 264 articles
- Spectral Graph Transducer. SVMlight can also train SVMs with cost models (see [Morik ... including text classification [Joachims, 1999c][Joachims, 1998a], image recognition tasks, bioinformatics and medical applications. Many...
- Referenced in 53 articles
- based frameworks, you can train object detection, object recognition, and image retrieval systems. Algorithms...
- Referenced in 7 articles
- reproduces multiplepoint statistics moments from training images; this allows modeling geologically realistic features, such ... storage of multiplepoint statistics moments from training images. SNESIM is able to simulate threedimensional models ... expense of data conditioning exactitude and/or training pattern reproduction quality. In this paper, the original ... multiple-point statistics moments from the training image and build the search tree, a simple...
- Referenced in 8 articles
- from both the available data and training images. This approach is data-driven and reconstructs ... realizations, while it only borrows from training images information that is not available...
- Referenced in 23 articles
- dataset is an image recognition dataset for developing unsupervised feature learning, deep learning, self-taught ... particular, each class has fewer labeled training examples than in CIFAR-10, but a very ... provided to learn image models prior to supervised training. The primary challenge is to make...
- Referenced in 5 articles
- approach to detect repeatable keypoints under drastic imaging changes of weather and lighting conditions ... identify good keypoint candidates in multiple training images taken from the same viewpoint. We then...
- Referenced in 5 articles
- form of patterns from a training image. Patterns are grouped into a pattern database...
- Referenced in 27 articles
- mixture model. We train the mixture model on a quarter million images. With ADVI...
- Referenced in 3 articles
- inferring the statistical model from a training image. This allows for a simpler quantification...
- Referenced in 16 articles
- ground truth images and the distribution of unregularized reconstructions. Once trained, the network is applied...
- Referenced in 56 articles
- artifacts due to image rotation are avoided and d) a new and larger clutter category ... Finally we use the clutter category to train an interest detector which rejects uninformative background...
- Referenced in 6 articles
- with distinct network architectures: (i) Image Classification: We train a 557-million-parameter AmoebaNet model...
- Referenced in 31 articles
- present a method for detecting objects in images using a single deep neural network ... single network. This makes SSD easy to train and straightforward to integrate into systems that ... while providing a unified framework for both training and inference. Compared to other single stage ... better accuracy, even with a smaller input image size. For 300×300 input, SSD achieves...