Na Lü

Yuxuan Yan, Na Lu, Ruofan Yan

Combining machine clustering with deep models has shown remarkable superiority in deep clustering. It modifies the data processing pipeline into two alternating phases: feature clustering and model training. However, such alternating schedule may lead to instability and computational burden issues. We propose a centerless clustering algorithm called Probability Aggregation Clustering (PAC) to proactively adapt deep learning technologies, enabling easy deployment in online deep clustering. PAC circumvents the cluster center and aligns the probability space and distribution space by formulating clustering as an optimization problem with a novel objective function. Based on the computation mechanism of the PAC, we propose a general online probability aggregation module to perform stable and flexible feature clustering over mini-batch data and further construct a deep visual clustering framework deep PAC (DPAC). Extensive experiments demonstrate that PAC has superior clustering robustness and performance and DPAC remarkably outperforms the state-of-the-art deep clustering methods.

Xu Niu, Na Lu, Huan Luo et al.

Phase synchrony information plays a crucial role in analyzing functional brain connectivity and identifying brain activities. A widely adopted feature extraction pipeline, composed of preprocessing, selection of EEG acquisition channels, and phase locking value (PLV) calculation, has achieved success in motor imagery classification (MI). However, this pipeline is manual and reliant on expert knowledge, limiting its convenience and adaptability to different application scenarios. Moreover, most studies have employed mediocre data-independent spatial filters to suppress noise, impeding the exploration of more significant phase synchronization phenomena. To address the issues, we propose the concept of phase synchrony component self-organization, which enables the adaptive learning of data-dependent spatial filters for automating both the preprocessing and channel selection procedures. Based on this concept, the first deep learning end-to-end network is developed, which directly extracts phase synchrony-based features from raw EEG signals and perform classification. The network learns optimal filters during training, which are obtained when the network achieves peak classification results. Extensive experiments have demonstrated that our network outperforms state-of-the-art methods. Remarkably, through the learned optimal filters, significant phase synchronization phenomena can be observed. Specifically, by calculating the PLV between a pair of signals extracted from each sample using two of the learned spatial filters, we have obtained an average PLV exceeding 0.87 across all tongue MI samples. This high PLV indicates a groundbreaking discovery in the synchrony pattern of tongue MI.

Hanxuan Wang, Na Lu, Qinyang Liu

Deep clustering outperforms conventional clustering by mutually promoting representation learning and cluster assignment. However, most existing deep clustering methods suffer from two major drawbacks. First, most cluster assignment methods are based on simple distance comparison and highly dependent on the target distribution generated by a handcrafted nonlinear mapping. These facts largely limit the possible performance that deep clustering methods can reach. Second, the clustering results can be easily guided towards wrong direction by the misassigned samples in each cluster. The existing deep clustering methods are incapable of discriminating such samples. To address these issues, a novel modular Self-Evolutionary Clustering (Self-EvoC) framework is constructed, which boosts the clustering performance by classification in a self-supervised manner. Fuzzy theory is used to score the sample membership with probability which evaluates the intermediate clustering result certainty of each sample. Based on which, the most reliable samples can be selected and augmented. The augmented data are employed to fine-tune an off-the-shelf deep network classifier with the labels from the clustering, which results in a model to generate the target distribution. The proposed framework can efficiently discriminate sample outliers and generate better target distribution with the assistance of self-supervised classifier. Extensive experiments indicate that the Self-EvoC remarkably outperforms state-of-the-art deep clustering methods on three benchmark datasets.

Zhiyan Cui, Na Lu

Most of the existing tracking methods based on CNN(convolutional neural networks) are too slow for real-time application despite the excellent tracking precision compared with the traditional ones. Moreover, neural networks are memory intensive which will take up lots of hardware resources. In this paper, a feature selection visual tracking algorithm combining CNN based MDNet(Multi-Domain Network) and RoIAlign was developed. We find that there is a lot of redundancy in feature maps from convolutional layers. So valid feature maps are selected by mutual information and others are abandoned which can reduce the complexity and computation of the network and do not affect the precision. The major problem of MDNet also lies in the time efficiency. Considering the computational complexity of MDNet is mainly caused by the large amount of convolution operations and fine-tuning of the network during tracking, a RoIAlign layer which could conduct the convolution over the whole image instead of each RoI is added to accelerate the convolution and a new strategy of fine-tuning the fully-connected layers is used to accelerate the update. With RoIAlign employed, the computation speed has been increased and it shows greater precision than RoIPool. Because RoIAlign can process float number coordinates by bilinear interpolation. These strategies can accelerate the processing, reduce the complexity with very low impact on precision and it can run at around 10 fps(while the speed of MDNet is about 1 fps). The proposed algorithm has been evaluated on a benchmark: OTB100, on which high precision and speed have been obtained.

Zhiyan Cui, Na Lu

Most of the existing tracking methods based on CNN(convolutional neural networks) are too slow for real-time application despite the excellent tracking precision compared with the traditional ones. In this paper, a fast dynamic visual tracking algorithm combining CNN based MDNet(Multi-Domain Network) and RoIAlign was developed. The major problem of MDNet also lies in the time efficiency. Considering the computational complexity of MDNet is mainly caused by the large amount of convolution operations and fine-tuning of the network during tracking, a RoIPool layer which could conduct the convolution over the whole image instead of each RoI is added to accelerate the convolution and a new strategy of fine-tuning the fully-connected layers is used to accelerate the update. With RoIPool employed, the computation speed has been increased but the tracking precision has dropped simultaneously. RoIPool could lose some positioning precision because it can not handle locations represented by floating numbers. So RoIAlign, instead of RoIPool, which can process floating numbers of locations by bilinear interpolation has been added to the network. The results show the target localization precision has been improved and it hardly increases the computational cost. These strategies can accelerate the processing and make it 7x faster than MDNet with very low impact on precision and it can run at around 7 fps. The proposed algorithm has been evaluated on two benchmarks: OTB100 and VOT2016, on which high precision and speed have been obtained. The influence of the network structure and training data are also discussed with experiments.

Na Lu, Hongyu Miao

Tree-structured data usually contain both topological and geometrical information, and are necessarily considered on manifold instead of Euclidean space for appropriate data parameterization and analysis. In this study, we propose a novel tree-structured data parameterization, called Topology-Attribute matrix (T-A matrix), so the data clustering task can be conducted on matrix manifold. We incorporate the structure constraints embedded in data into the negative matrix factorization method to determine meta-trees from the T-A matrix, and the signature vector of each single tree can then be extracted by meta-tree decomposition. The meta-tree space turns out to be a cone space, in which we explore the distance metric and implement the clustering algorithm based on the concepts like Fréchet mean. Finally, the T-A matrix based clustering (TAMBAC) framework is evaluated and compared using both simulated data and real retinal images to illustrate its efficiency and accuracy.