Depression has become a major global public health challenge. Social media texts reflect users’ emotional and psychological states, providing a valuable data source for depression detection through natural language a...
详细信息
To overcome image degradation under the conditions of haze,and fog,a method using conditional generative adversarial defogging algorithm based on polarization characteristic is *** original images with different polar...
详细信息
ISBN:
(纸本)9781665431293
To overcome image degradation under the conditions of haze,and fog,a method using conditional generative adversarial defogging algorithm based on polarization characteristic is *** original images with different polarization angles were obtained from the original image,and then the polarization images were characterized using Stokes *** the relationship between Stokes vector and polarization image,each polarization image with different angles is input into the same network to extract *** the same time,the enhanced network is used to extract the characteristics of the fog area in the polarization image,and the polarization information is extracted by layer jump connection,which is fused with the image features of different *** constructing the loss function and obtaining the optimal solution,the fog-free image is finally *** results show that a clear image can be reconstructed in fog situations by using conditional generative adversarial network-based on polarization characteristics,and the structural similarity is improved by about 10%,the peak signal to noise ratio is increased by about 0.5.
Diffusion models are powerful generative models, and this capability can also be applied to discrimination. The inner activations of a pre-trained diffusion model can serve as features for discriminative tasks, namely...
ISBN:
(纸本)9798331314385
Diffusion models are powerful generative models, and this capability can also be applied to discrimination. The inner activations of a pre-trained diffusion model can serve as features for discriminative tasks, namely, diffusion feature. We discover that diffusion feature has been hindered by a hidden yet universal phenomenon that we call content shift. To be specific, there are content differences between features and the input image, such as the exact shape of a certain object. We locate the cause of content shift as one inherent characteristic of diffusion models, which suggests the broad existence of this phenomenon in diffusion feature. Further empirical study also indicates that its negative impact is not negligible even when content shift is not visually perceivable. Hence, we propose to suppress content shift to enhance the overall quality of diffusion features. Specifically, content shift is related to the information drift during the process of recovering an image from the noisy input, pointing out the possibility of turning off-the-shelf generation techniques into tools for content shift suppression. We further propose a practical guideline named GATE to efficiently evaluate the potential benefit of a technique and provide an implementation of our methodology. Despite the simplicity, the proposed approach has achieved superior results on various tasks and datasets, validating its potential as a generic booster for diffusion features. Our code is available at https://***/Darkbblue/diffusion-content-shift.
Compound property assays are an important part of drug development, but incomplete data may occur for a variety of reasons. To deal with these incomplete data and improve the success rate of drug development, research...
详细信息
Class-incremental learning (CIL) enables models to learn new classes progressively while preserving knowledge of previously learned ones. Recent advances in this field have shifted towards parameter-efficient fine-tun...
详细信息
We present a no-reference image-quality - assessment algorithm based on active reasoning module. This algorithm has three modules: the feature extraction module, the active reasoning module, and the quality assessment...
详细信息
As an extended rough set model, the probabilistic rough sets can effectively process data with noise. How to effectively calculate three approximation regions, is a crucial issue of probability rough sets. Existing ca...
详细信息
Microphone array techniques are widely used in sound source localization and smart city acoustic-based traffic monitoring, but these applications face significant challenges due to the scarcity of labeled real-world t...
详细信息
ISBN:
(数字)9798350368741
ISBN:
(纸本)9798350368758
Microphone array techniques are widely used in sound source localization and smart city acoustic-based traffic monitoring, but these applications face significant challenges due to the scarcity of labeled real-world traffic audio data and the complexity and diversity of application scenarios. The DCASE Challenge’s Task 10 focuses on using multi-channel audio signals to count vehicles (cars or commercial vehicles) and identify their directions (left-to-right or vice versa). In this paper, we propose a graph-enhanced dual-stream feature fusion network (GEDF-Net) for acoustic traffic monitoring, which simultaneously considers vehicle type and direction to improve detection. We propose a graph-enhanced dual-stream feature fusion strategy which consists of a vehicle type feature extraction (VTFE) branch, a vehicle direction feature extraction (VDFE) branch, and a frame-level feature fusion module to combine the type and direction feature for enhanced performance. A pre-trained model (PANNs) is used in the VTFE branch to mitigate data scarcity and enhance the type features, followed by a graph attention mechanism to exploit temporal relationships and highlight important audio events within these features. The frame-level fusion of direction and type features enables fine-grained feature representation, resulting in better detection performance. Experiments demonstrate the effectiveness of our proposed method. GEDF-Net is our submission that achieved 1st place in the DCASE 2024 Challenge Task 10.
At present, Internet of Things (IoT) testing instruments and meters are unable to identify the communication protocol of communication data when all characteristic parameters of communication data are unknown. Regardi...
详细信息
ISBN:
(数字)9798350377675
ISBN:
(纸本)9798350377682
At present, Internet of Things (IoT) testing instruments and meters are unable to identify the communication protocol of communication data when all characteristic parameters of communication data are unknown. Regarding this problem, this paper utilizes the K-Nearest Neighbors (KNN) model in machine learning, designs communication protocol identification methods and proposes a communication protocol identification method based on time-frequency domain combination (CPI- TFC). First, we generate time-domain waveforms and frequency-domain waveforms from sampling communication data sequences, and attempt only time-domain identification method and only frequency-domain identification method based on KNN model. Then, we compare the results, design a communication protocol identification method based on the combination of time-domain identification and frequency domain identification, and propose CPI-TFC. Experimental results show that CPI-TFC can achieve higher identification accuracy than only time-domain identification method and only frequency-domain identification method, so it can effectively identify the communication protocol for a IoT communication data. Moreover, CPI- TFC can provide prior information for subsequent configuring parameters in IoT testing instruments, improve the parameter configuration efficiency.
High-dynamic scene optical flow is a challenging task, which suffers spatial blur and temporal discontinuous motion due to large displacement in frame imaging, thus deteriorating the spatiotemporal feature of optical ...
详细信息
暂无评论