In this study, an advanced Kinect sensor was adopted to acquire infrared radiation (IR) images for liveness detection. The proposed liveness detection method based on infrared radiation (IR) images can deal with face spoofs. Face pictures were acquired by a Kinect camera and converted into IR images. Feature extraction and classification were carried out by a deep neural network to distinguish between real individuals and face spoofs. IR images collected by the Kinect camera have depth information. Therefore, the IR pixels from live images have an evident hierarchical structure, while those from photos or videos have no evident hierarchical feature. Accordingly, two types of IR images were learned through the deep network to realize the identification of whether images were from live individuals. In comparison with other liveness detection cross-databases, our recognition accuracy was 99.8% and better than other algorithms. FaceNet is a face recognition model, and it is robust to occlusion, blur, illumination, and steering. We combined the liveness detection and FaceNet model for identity authentication. For improving the application of the authentication approach, we proposed two improved ways to run the FaceNet model. Experimental results showed that the combination of the proposed liveness detection and improved face recognition had a good recognition effect and can be used for identity authentication.

Automatic text summarization is a method used to compress documents while preserving the main idea of the original text, including extractive summarization and abstractive summarization. Extractive text summarization extracts important sentences from the original document to serve as the summary. The document representation method is crucial for the quality of the generated summarization. To effectively represent the document, we propose a hierarchical document representation model Long-Trans-Extr for Extractive Summarization, which uses Longformer as the sentence encoder and Transformer as the document encoder. The advantage of Longformer as sentence encoder is that the model can input long document up to 4096 tokens with adding relative a little calculation. The proposed model Long-Trans-Extr is evaluated on three benchmark datasets: CNN (Cable News Network), DailyMail, and the combined CNN/DailyMail. It achieves 43.78 (Rouge-1) and 39.71 (Rouge-L) on CNN/DailyMail and 33.75 (Rouge-1), 13.11 (Rouge-2), and 30.44 (Rouge-L) on the CNN datasets. They are very competitive results, and furthermore, they show that our model has better performance on long documents, such as the CNN corpus.