Model drifting is one of the knotty problems that seriously restricts the accuracy of discriminative trackers in visual tracking. Most existing works usually focus on improving the robustness of the target appearance model. However, they are prone to suffer from model drifting due to the inappropriate model updates during the tracking-by-detection. In this paper, we propose a novel update-pacing framework to suppress the occurrence of model drifting in visual tracking. Specifically, the proposed framework first initializes an ensemble of trackers, each of which updates the model in a different update interval. Once the forward tracking trajectory of each tracker is determined, the backward trajectory will also be generated by the current model to measure the difference with the forward one, and the tracker with the smallest deviation score will be selected as the most robust tracker for the remaining tracking. By performing such self-examination on trajectory pairs, the framework can effectively preserve the temporal context consistency of sequential frames to avoid learning corrupted information. To further improve the performance of the proposed method, a multi-feature extension framework is also proposed to incorporate multiple features into the ensemble of the trackers. The extensive experimental results obtained on large-scale object tracking benchmarks demonstrate that the proposed framework significantly increases the accuracy and robustness of the underlying base trackers, such as DSST, Struck, KCF, and CT, and achieves superior performance compared with the state-of-the-art methods without using deep models.

Light field (LF) photography is an emerging paradigm for capturing more immersive representations of the real-world. However, arising from the inherent trade-off between the angular and spatial dimensions, the spatial resolution of LF images captured by commercial micro-lens based LF cameras are significantly constrained. In this paper, we propose effective and efficient end-to-end convolutional neural network models for spatially super-resolving LF images. Specifically, the proposed models have an hourglass shape, which allows feature extraction to be performed at the low resolution level to save both computational and memory costs. To fully make use of the four-dimensional (4-D) structure information of LF data in both spatial and angular domains, we propose to use 4-D convolution to characterize the relationship among pixels. Moreover, as an approximation of 4-D convolution, we also propose to use spatialangular separable (SAS) convolutions for more computationallyand memory-efficient extraction of spatial-angular joint features. Extensive experimental results on 57 test LF images with various challenging natural scenes show significant advantages from the proposed models over state-of-the-art methods. That is, an average PSNR gain of more than 3.0 dB and better visual quality are achieved, and our methods preserve the LF structure of the super-resolved LF images better, which is highly desirable for subsequent applications. In addition, the SAS convolutionbased model can achieve 3× speed up with only negligible reconstruction quality decrease when compared with the 4-D convolution-based one. The source code of our method is online available at https://github.com/spatialsr/DeepLightFieldSSR.

Automated person re-identification is a challenging research problem that has many real-world applications, especially in video surveillance. While many recent studies have been focusing on solving the person re-identification problem using full-scale images or video footages, little work has been done to solve the person re-identification problem in a top-down context. In this work, we propose a solution to the top-down reidentification problem that uses the Siamese architecture in conjunction with Convolutional Neural Networks. In our approach, a pair of top-down images is distinguished by a single Siamese network, which is trained to predict the similarity, or a distance between two input images. Experiments have shown that once the model is properly trained, it is able to achieve one-shot, top-down re-identification by learning unseen classes of person in real-time.

This paper proposes the use of Stacked Denoising Autoencoder to predict the direction of movement of stock indexes based on the historical and volume data of the underlying stocks. The Stacked Denoising Autoencoder is a deep learning method widely used in the field of computer vision which is capable of learning a compact feature representation of the data for stock index prediction. The Hybrid Gravitational Search Algorithm, a swam intelligence based algorithm, is proposed to optimise the hyper-parameters of the deep network which mitigates the hyper-parameter tuning problem of the network.

Recognizing human faces is one of the most popular problems in the field of pattern recognition. Many approaches and methods have been tested and applied on the topic, especially neural networks. This paper proposed a new loss layer that can be replaced at the bottom of a neural network architecture in terms of face recognition, called constrained triplet loss layer (CTLL). In order to make more confident predictions and classifications, this loss layer helps the deep learning model to specify further distinguishable clusters between different people (classes) by placing extra constraints on images of the same person (intra-person) while putting margins on images of a different person (inter-person). This proposed constrained triplet loss layer improved the recognition accuracy on faces by 2%.