This work presents an unsupervised learning based approach to the ubiquitous computer vision problem of image matching. We start from the insight that the problem of frame-interpolation implicitly solves for inter-frame correspondences. This permits the application of analysisby-synthesis: we firstly train and apply a Convolutional Neural Network for frame-interpolation, then obtain correspondences by inverting the learned CNN. The key benefit behind this strategy is that the CNN for frame-interpolation can be trained in an unsupervised manner by exploiting the temporal coherency that is naturally contained in real-world video sequences. The present model therefore learns image matching by simply "watching videos". Besides a promise to be more generally applicable, the presented approach achieves surprising performance comparable to traditional empirically designed methods.

No abstract

No abstract

In this paper, aiming at the small target detection problem in the infrared image sequence, we propose a small target detection method based on maximum likelihood estimation and NNLoG spot detection operator. Compared with the traditional method, our proposed method can partially solve the nonlinear motion of the small target in image sequence. The real target trajectory is approximated by polynomial to enhance the signal to noise ratio of target. To validate the proposed method, we create eight experiments to simulate. The experiment result shows that our method is very valuable for small target detection.