PupilNet: Principled Face Alignment with Recurrent Attention – In this paper, we propose an attention-based model for visual attention. Previous work explicitly uses the attention mechanism to learn attention maps instead of a feature. However, previous studies focused on the visual attention mechanism which was not explored. Here, we explore the visual attention mechanism using a feature. A key assumption in previous attention-based approaches is that visual attention consists of learning two representations of visual features, and each of these representations may be used in different tasks. We propose a novel visual attention mechanism that learns attention maps by visualizing the task at hand and using a deep learning algorithm to adaptively update the representations of visual features. Experimental results using a new state-of-the-art visual attention system, the CNN-D+R-DI, demonstrate that the proposed method achieves competitive recognition rate of 90.9 per cent (95%) on the MNIST dataset.
We present a novel toolkit for machine translation. Our goal is to provide a machine translation system with the ability to extract, encode, and classify text with the ability to process annotations from different languages. We are aiming to provide a framework for automatic classification, a language model based on sentence generation and data interpretation, and a model that can incorporate the human annotation process. Our system achieves excellent results including a recognition rate of 95.7% on TREC and 80.5% on JAVA.
Learning Action Proposals from Unconstrained Videos
Dependency-Based Deep Recurrent Models for Answer Recommendation
PupilNet: Principled Face Alignment with Recurrent Attention
Deep Spatio-Temporal Learning of Motion Representations
Learning Text and Image Descriptions from Large Scale Video Annotations with Semi-supervised LearningWe present a novel toolkit for machine translation. Our goal is to provide a machine translation system with the ability to extract, encode, and classify text with the ability to process annotations from different languages. We are aiming to provide a framework for automatic classification, a language model based on sentence generation and data interpretation, and a model that can incorporate the human annotation process. Our system achieves excellent results including a recognition rate of 95.7% on TREC and 80.5% on JAVA.