Cross spectral stereo matching is a challenging task due to different spectral properties causing unreliable results in correspondence estimation. In this paper, we propose joint disparity estimation and pseudo near infrared (NIR) generation from cross spectral image pairs. To bridge the spectral gap between paired images, we adopt differential map operations and non-local blocks to improve the local attention and global attention of the network. The proposed network is based on unsupervised learning that consists of one encoder and two decoders, which performs both spectral translation and disparity estimation. For cooperative learning, we use difference map operation to connect two decoders, thus improving the inference ability of the decoder in regions even with large spectral differences. Experimental results show that the proposed network achieves good performance in cross spectral stereo matching for unreliable regions such as shadows and glasses. Moreover, the proposed network generates pseudo NIR images nearly the same as the ground truth even in the regions with large spectral difference. Besides, we achieve real-time speed of 27 FPS for 582×429 image pairs on RTX 2060 6G GPU due to the low computational complexity.

Auto-Generated Test Paper (AGTP) has been deeply studied for many years, however, it is still a difficult problem and the certainty to access the best test paper (TP) is not guaranteed yet. In this paper, we put forward a method for AGTP based on knowledge embedding, which makes AGTP easier and faster. The knowledge to be embedded is studied and the mechanism behind it is analyzed. The embedded knowledge in this paper is from both the constraints of TP and the information of question repository (QR). The experiments validated the proposed method and found it is not only faster but also costs less computational resources to access the best TP than other method, such as evolutionary algorithm. What impressed is that the cost time to access the optimum does not rapidly increase with the size of QR. The knowledge plays the important role in AGTP, especially to efficiently improve the performance of the algorithms.