Wei Shen

TL;DR: A novel method that can adapt a pre-trained neural network to novel categories by directly predicting the parameters from the activations is proposed, which achieves the state-of-the-art classification accuracy on Novel categories by a significant margin while keeping comparable performance on the large-scale categories.

TL;DR: This work shows that contour detection accuracy can be improved by instead making the use of the deep features learned from convolutional neural networks (CNNs), while rather than using the networks as a blackbox feature extractor, it customize the training strategy by partitioning contour (positive) data into subclasses and fitting each subclass by different model parameters.

TL;DR: A novel approach for text detection in natural images that consistently achieves the state-of-the-art performance on three text detection benchmarks: MSRA-TD500, I CDAR2015 and ICDAR2013.

TL;DR: In this article, a Fully Convolutional Network (FCN) model is trained to predict the salient map of text regions in a holistic manner, and text line hypotheses are estimated by combining the saliency map and character components.

TL;DR: Deep Co-training as discussed by the authors exploits adversarial examples to encourage view difference, in order to prevent the networks from collapsing into each other and provide complementary information about the data, which is necessary for the co-training framework to achieve good results.