In this paper, we present Multi-scale Feature Aggregation Conformer (MFA-Conformer), an easy-to-implement, simple but effective backbone for automatic speaker verification based on the Convolution-augmented Transformer (Conformer). The architecture of the MFA-Conformer is inspired by recent state-of-the-art models in speech recognition and speaker verification. Firstly, we introduce a convolution sub-sampling layer to decrease the computational cost of the model. Secondly, we adopt Conformer blocks which combine Transformers and convolution neural networks (CNNs) to capture global and local features effectively. Finally, the output feature maps from all Conformer blocks are concatenated to aggregate multi-scale representations before final pooling. We evaluate the MFA-Conformer on the widely used benchmarks. The best system obtains 0.64%, 1.29% and 1.63% EER on VoxCeleb1-O, SITW.Dev, and SITW.Eval set, respectively. MFA-Conformer significantly outperforms the popular ECAPA-TDNN systems in both recognition performance and inference speed. Last but not the least, the ablation studies clearly demonstrate that the combination of global and local feature learning can lead to robust and accurate speaker embedding extraction. We will release the code for future works to do comparison.
Knowledge Distillation (KD) is a widely-used technology to inherit information from cumbersome teacher models to compact student models, consequently realizing model compression and acceleration. Compared with image classification, object detection is a more complex task, and designing specific KD methods for object detection is non-trivial. In this work, we elaborately study the behaviour difference between the teacher and student detection models, and obtain two intriguing observations: First, the teacher and student rank their detected candidate boxes quite differently, which results in their precision discrepancy. Second, there is a considerable gap between the feature response differences and prediction differences between teacher and student, indicating that equally imitating all the feature maps of the teacher is the sub-optimal choice for improving the student's accuracy. Based on the two observations, we propose Rank Mimicking (RM) and Prediction-guided Feature Imitation (PFI) for distilling one-stage detectors, respectively. RM takes the rank of candidate boxes from teachers as a new form of knowledge to distill, which consistently outperforms the traditional soft label distillation. PFI attempts to correlate feature differences with prediction differences, making feature imitation directly help to improve the student's accuracy. On MS COCO and PASCAL VOC benchmarks, extensive experiments are conducted on various detectors with different backbones to validate the effectiveness of our method. Specifically, RetinaNet with ResNet50 achieves 40.4% mAP in MS COCO, which is 3.5% higher than its baseline, and also outperforms previous KD methods.
As an effective algorithm for solving complex optimization problems, artificial bee colony (ABC) algorithm has shown to be competitive, but the same as other population-based algorithms, it is poor at balancing the abilities of global searching in the whole solution space (named as exploration) and quick searching in local solution space which is defined as exploitation. For improving the performance of ABC, an adaptive group collaborative ABC (AgABC) algorithm is introduced where the population in different phases is divided to specific groups and different search strategies with different abilities are assigned to the members in groups, and the member or strategy which obtains the best solution will be employed for further searching. Experimental results on benchmark functions show that the proposed algorithm with dynamic mechanism is superior to other algorithms in searching accuracy and stability. Furthermore, numerical experiments show that the proposed method can generate the optimal solution for the complex scheduling problem.
The operating state of bearing directly affects the performance of rotating machinery and how to accurately and decisively extract features from the original vibration signal and recognize the faulty parts as early as possible is very critical. In this study, the one-dimensional ternary model which has been proved to be an effective statistical method in feature selection is introduced and shapelets transformation is proposed to calculate the parameter of it which is also the standard deviation of the transformed shaplets that is usually selected by trial and error. Moreover, XGBoost is used to recognize the faults from the obtained features, and an improved artificial bee colony algorithm(ABC) where the evolution is guided by the importance indices of different search space is proposed to optimize the parameters of XGBoost. Here the value of importance index is related to the probability of optimal solutions in certain space, thus the problem of easily falling into local optimality in traditional ABC could be avoided.The experimental results based on the failure vibration signal samples show that the average accuracy of fault signal recognition can reach 97% which is much higher than the ones corresponding to other extraction strategies, thus the ability of extraction could be improved. And with the improved artificial bee colony algorithm which is used to optimize the parameters of XGBoost, the classification accuracy could be improved from 97.02% to about 98.60% compared with the traditional classification strategy
Video-based person re-identification (re-ID) is an important technique in visual surveillance systems which aims to match video snippets of people captured by different cameras. Existing methods are mostly based on convolutional neural networks (CNNs), whose building blocks either process local neighbor pixels at a time, or, when 3D convolutions are used to model temporal information, suffer from the misalignment problem caused by person movement. In this paper, we propose to overcome the limitations of normal convolutions with a human-oriented graph method. Specifically, features located at person joint keypoints are extracted and connected as a spatial-temporal graph. These keypoint features are then updated by message passing from their connected nodes with a graph convolutional network (GCN). During training, the GCN can be attached to any CNN-based person re-ID model to assist representation learning on feature maps, whilst it can be dropped after training for better inference speed. Our method brings significant improvements over the CNN-based baseline model on the MARS dataset with generated person keypoints and a newly annotated dataset: PoseTrackReID. It also defines a new state-of-the-art method in terms of top-1 accuracy and mean average precision in comparison to prior works.
In order to robustly deploy object detectors across a wide range of scenarios, they should be adaptable to shifts in the input distribution without the need to constantly annotate new data. This has motivated research in Unsupervised Domain Adaptation (UDA) algorithms for detection. UDA methods learn to adapt from labeled source domains to unlabeled target domains, by inducing alignment between detector features from source and target domains. Yet, there is no consensus on what features to align and how to do the alignment. In our work, we propose a framework that generalizes the different components commonly used by UDA methods laying the ground for an in-depth analysis of the UDA design space. Specifically, we propose a novel UDA algorithm, ViSGA, a direct implementation of our framework, that leverages the best design choices and introduces a simple but effective method to aggregate features at instance-level based on visual similarity before inducing group alignment via adversarial training. We show that both similarity-based grouping and adversarial training allows our model to focus on coarsely aligning feature groups, without being forced to match all instances across loosely aligned domains. Finally, we examine the applicability of ViSGA to the setting where labeled data are gathered from different sources. Experiments show that not only our method outperforms previous single-source approaches on Sim2Real and Adverse Weather, but also generalizes well to the multi-source setting.
Automatic speaker verification (ASV), one of the most important technology for biometric identification, has been widely adopted in security-critical applications, including transaction authentication and access control. However, previous work has shown that ASV is seriously vulnerable to recently emerged adversarial attacks, yet effective countermeasures against them are limited. In this paper, we adopt neural vocoders to spot adversarial samples for ASV. We use the neural vocoder to re-synthesize audio and find that the difference between the ASV scores for the original and re-synthesized audio is a good indicator for discrimination between genuine and adversarial samples. This effort is, to the best of our knowledge, among the first to pursue such a technical direction for detecting adversarial samples for ASV, and hence there is a lack of established baselines for comparison. Consequently, we implement the Griffin-Lim algorithm as the detection baseline. The proposed approach achieves effective detection performance that outperforms all the baselines in all the settings. We also show that the neural vocoder adopted in the detection framework is dataset-independent. Our codes will be made open-source for future works to do comparison.
Current datasets for video-based person re-identification (re-ID) do not include structural knowledge in form of human pose annotations for the persons of interest. Nonetheless, pose information is very helpful to disentangle useful feature information from background or occlusion noise. Especially real-world scenarios, such as surveillance, contain a lot of occlusions in human crowds or by obstacles. On the other hand, video-based person re-ID can benefit other tasks such as multi-person pose tracking in terms of robust feature matching. For that reason, we present PoseTrackReID, a large-scale dataset for multi-person pose tracking and video-based person re-ID. With PoseTrackReID, we want to bridge the gap between person re-ID and multi-person pose tracking. Additionally, this dataset provides a good benchmark for current state-of-the-art methods on multi-frame person re-ID.
In this work, we tackle the problem of person search, which is a challenging task consisted of pedestrian detection and person re-identification~(re-ID). Instead of sharing representations in a single joint model, we find that separating detector and re-ID feature extraction yields better performance. In order to extract more representative features for each identity, we segment out the foreground person from the original image patch. We propose a simple yet effective re-ID method, which models foreground person and original image patches individually, and obtains enriched representations from two separate CNN streams. From the experiments on two standard person search benchmarks of CUHK-SYSU and PRW, we achieve mAP of $83.0\%$ and $32.6\%$ respectively, surpassing the state of the art by a large margin (more than 5pp).
Person re-identification aims to match a person's identity across multiple camera streams. Deep neural networks have been successfully applied to the challenging person re-identification task. One remarkable bottleneck is that the existing deep models are data hungry and require large amounts of labeled training data. Acquiring manual annotations for pedestrian identity matchings in large-scale surveillance camera installations is a highly cumbersome task. Here, we propose the first semi-supervised approach that performs pseudo-labeling by considering complex relationships between unlabeled and labeled training samples in the feature space. Our approach first approximates the actual data manifold by learning a generative model via adversarial training. Given the trained model, data augmentation can be performed by generating new synthetic data samples which are unlabeled. An open research problem is how to effectively use this additional data for improved feature learning. To this end, this work proposes a novel Feature Affinity based Pseudo-Labeling (FAPL) approach with two possible label encodings under a unified setting. Our approach measures the affinity of unlabeled samples with the underlying clusters of labeled data samples using the intermediate feature representations from deep networks. FAPL trains with the joint supervision of cross-entropy loss together with a center regularization term, which not only ensures discriminative feature representation learning but also simultaneously predicts pseudo-labels for unlabeled data. Our extensive experiments on two standard large-scale datasets, Market-1501 and DukeMTMC-reID, demonstrate significant performance boosts over closely related competitors and outperforms state-of-the-art person re-identification techniques in most cases.