An Uncertainty-based Neural Network for Explainable Trajectory Segmentation

As a variant task of time-series segmentation, trajectory segmentation is a key task in the applications of transportation pattern recognition and traffic analysis. However, segmenting trajectory is faced with challenges of implicit patterns and sparse results. Although deep neural networks have tremendous advantages in terms of high-level feature learning performance, deploying as a blackbox seriously limits the real-world applications. Providing explainable segmentations has significance for result evaluation and decision making. Thus, in this article, we address trajectory segmentation by proposing a Bayesian Encoder-Decoder Network (BED-Net) to provide accurate detection with explainability and references for the following active-learning procedures. BED-Net consists of a segmentation module based on Monte Carlo dropout and an explanation module based on uncertainty learning that provides results evaluation and visualization. Experimental results on both benchmark and real-world datasets indicate that BED-Net outperforms the rival methods and offers excellent explainability in the applications of trajectory segmentation.

Spatio-Temporal Event Forecasting Using Incremental Multi-Source Feature Learning

The forecasting of significant societal events such as civil unrest and economic crisis is an interesting and challenging problem which requires both timeliness, precision, and comprehensiveness. Significant societal events are influenced and indicated jointly by multiple aspects of a society, including its economics, politics, and culture. Traditional forecasting methods based on a single data source find it hard to cover all these aspects comprehensively, thus limiting model performance. Multi-source event forecasting has proven promising but still suffers from several challenges, including (1) geographical hierarchies in multi-source data features, (2) hierarchical missing values, (3) characterization of structured feature sparsity, and (4) difficulty in model’s online update with incomplete multiple sources. This article proposes a novel feature learning model that concurrently addresses all the above challenges. Specifically, given multi-source data from different geographical levels, we design a new forecasting model by characterizing the lower-level features’ dependence on higher-level features. To handle the correlations amidst structured feature sets and deal with missing values among the coupled features, we propose a novel feature learning model based on an N th-order strong hierarchy and fused-overlapping group Lasso. An efficient algorithm is developed to optimize model parameters and ensure global optima. More importantly, to enable the model update in real time, the online learning algorithm is formulated and active set techniques are leveraged to resolve the crucial challenge when new patterns of missing features appear in real time. Extensive experiments on 10 datasets in different domains demonstrate the effectiveness and efficiency of the proposed models.

A Plant Disease Recognition Method Based on Fusion of Images and Graph Structure Text

The disease image recognition models based on deep learning have achieved relative success under limited and restricted conditions, but such models are generally subjected to the shortcoming of weak robustness. The model accuracy would decrease obviously when recognizing disease images with complex backgrounds under field conditions. Moreover, most of the models based on deep learning only involve characterization learning on visual information in the image form, while the expression of other modal information rather than the image form is often ignored. The present study targeted the main invasive diseases in tomato and cucumber as the research object. Firstly, in response to the problem of weak robustness, a feature decomposition and recombination method was proposed to allow the model to learn image features at different granularities so as to accurately recognize different test images. Secondly, by extracting the disease feature words from the disease text description information composed of continuous vectors and recombining them into the disease graph structure text, the graph convolutional neural network (GCN) was then applied for feature learning. Finally, a vegetable disease recognition model based on the fusion of images and graph structure text was constructed. The results show that the recognition accuracy, precision, sensitivity, and specificity of the proposed model were 97.62, 92.81, 98.54, and 93.57%, respectively. This study improved the model robustness to a certain extent, and provides ideas and references for the research on the fusion method of image information and graph structure information in disease recognition.

3D object detection combining semantic and geometric features from point clouds

Background: 3D object detection based on point clouds in road scenes has attracted much attention recently. The voxel-based methods voxelize the scene to regular grids, which can be processed with the advanced feature learning frameworks based on convolutional layers for semantic feature learning. The point-based methods can extract the geometric feature of the point due to the coordinate reservations. The combination of the two is effective for 3D object detection. However, the current methods use a voxel-based detection head with anchors for classification and localization. Although the preset anchors cover the entire scene, it is not suitable for detection tasks with larger scenes and multiple categories of objects, due to the limitation of the voxel size. Additionally, the misalignment between the predicted confidence and proposals in the Regions of the Interest (ROI) selection bring obstacles to 3D object detection. Methods: We investigate the combination of voxel-based methods and point-based methods for 3D object detection. Additionally, a voxel-to-point module that captures semantic and geometric features is proposed in the paper. The voxel-to-point module is conducive to the detection of small-size objects and avoids the presets of anchors in the inference stage. Moreover, a confidence adjustment module with the center-boundary-aware confidence attention is proposed to solve the misalignment between the predicted confidence and proposals in the regions of the interest selection. Results: The proposed method has achieved state-of-the-art results for 3D object detection in the Karlsruhe Institute of Technology and Toyota Technological Institute (KITTI) object detection dataset. Actually, as of September 19, 2021, our method ranked 1st in the 3D and Bird Eyes View (BEV) detection of cyclists tagged with difficulty level ‘easy’, and ranked 2nd in the 3D detection of cyclists tagged with ‘moderate’. Conclusions: We propose an end-to-end two-stage 3D object detector with voxel-to-point module and confidence adjustment module.

Rethinking Pooling Operation for Liver and Liver-Tumor Segmentations

Deep convolutional neural networks (DCNNs) have been widely used in medical image segmentation due to their excellent feature learning ability. In these DCNNs, the pooling operation is usually used for image down-sampling, which can gradually reduce the image resolution and thus expands the receptive field of convolution kernel. Although the pooling operation has the above advantages, it inevitably causes information loss during the down-sampling of the pooling process. This paper proposes an effective weighted pooling operation to address the problem of information loss. First, we set up a pooling window with learnable parameters, and then update these parameters during the training process. Secondly, we use weighted pooling to improve the full-scale skip connection and enhance the multi-scale feature fusion. We evaluated weighted pooling on two public benchmark datasets, the LiTS2017 and the CHAOS. The experimental results show that the proposed weighted pooling operation effectively improve network performance and improve the accuracy of liver and liver-tumor segmentation.

Semantic-guided autoencoder adversarial hashing for large-scale cross-modal retrieval

With the vigorous development of mobile Internet technology and the popularization of smart devices, while the amount of multimedia data has exploded, its forms have become more and more diversified. People’s demand for information is no longer satisfied with single-modal data retrieval, and cross-modal retrieval has become a research hotspot in recent years. Due to the strong feature learning ability of deep learning, cross-modal deep hashing has been extensively studied. However, the similarity of different modalities is difficult to measure directly because of the different distribution and representation of cross-modal. Therefore, it is urgent to eliminate the modal gap and improve retrieval accuracy. Some previous research work has introduced GANs in cross-modal hashing to reduce semantic differences between different modalities. However, most of the existing GAN-based cross-modal hashing methods have some issues such as network training is unstable and gradient disappears, which affect the elimination of modal differences. To solve this issue, this paper proposed a novel Semantic-guided Autoencoder Adversarial Hashing method for cross-modal retrieval (SAAH). First of all, two kinds of adversarial autoencoder networks, under the guidance of semantic multi-labels, maximize the semantic relevance of instances and maintain the immutability of cross-modal. Secondly, under the supervision of semantics, the adversarial module guides the feature learning process and maintains the modality relations. In addition, to maintain the inter-modal correlation of all similar pairs, this paper use two types of loss functions to maintain the similarity. To verify the effectiveness of our proposed method, sufficient experiments were conducted on three widely used cross-modal datasets (MIRFLICKR, NUS-WIDE and MS COCO), and compared with several representatives advanced cross-modal retrieval methods, SAAH achieved leading retrieval performance.