arXiv subjects

Youshan Zhang

Publications and source records attributed to Youshan Zhang.

At least 37 records · Page 2Linked to original sources

arXiv2024

Vision Transformer Segmentation for Visual Bird Sound Denoising

Audio denoising, especially in the context of bird sounds, remains a challenging task due to persistent residual noise. Traditional and deep learning methods often struggle with artificial or low-frequency noise. In this work, we propose ViTVS, a novel approach that leverages the power of the vision transformer (ViT) architecture. ViTVS adeptly combines segmentation techniques to disentangle clean audio from complex signal mixtures. Our key contributions encompass the development of ViTVS, introducing comprehensive, long-range, and multi-scale representations. These contributions directly tackle the limitations inherent in conventional approaches. Extensive experiments demonstrate that ViTVS outperforms state-of-the-art methods, positioning it as a benchmark solution for real-world bird sound denoising applications. Source code is available at: https://github.com/aiai-4/ViVTS.

cs.SD↗

arXiv2023

DPATD: Dual-Phase Audio Transformer for Denoising

Recent high-performance transformer-based speech enhancement models demonstrate that time domain methods could achieve similar performance as time-frequency domain methods. However, time-domain speech enhancement systems typically receive input audio sequences consisting of a large number of time steps, making it challenging to model extremely long sequences and train models to perform adequately. In this paper, we utilize smaller audio chunks as input to achieve efficient utilization of audio information to address the above challenges. We propose a dual-phase audio transformer for denoising (DPATD), a novel model to organize transformer layers in a deep structure to learn clean audio sequences for denoising. DPATD splits the audio input into smaller chunks, where the input length can be proportional to the square root of the original sequence length. Our memory-compressed explainable attention is efficient and converges faster compared to the frequently used self-attention module. Extensive experiments demonstrate that our model outperforms state-of-the-art methods.

cs.SD↗

arXiv2023

DCHT: Deep Complex Hybrid Transformer for Speech Enhancement

Most of the current deep learning-based approaches for speech enhancement only operate in the spectrogram or waveform domain. Although a cross-domain transformer combining waveform- and spectrogram-domain inputs has been proposed, its performance can be further improved. In this paper, we present a novel deep complex hybrid transformer that integrates both spectrogram and waveform domains approaches to improve the performance of speech enhancement. The proposed model consists of two parts: a complex Swin-Unet in the spectrogram domain and a dual-path transformer network (DPTnet) in the waveform domain. We first construct a complex Swin-Unet network in the spectrogram domain and perform speech enhancement in the complex audio spectrum. We then introduce improved DPT by adding memory-compressed attention. Our model is capable of learning multi-domain features to reduce existing noise on different domains in a complementary way. The experimental results on the BirdSoundsDenoising dataset and the VCTK+DEMAND dataset indicate that our method can achieve better performance compared to state-of-the-art methods.

cs.SD↗

arXiv2023

Deep Feature Registration for Unsupervised Domain Adaptation

While unsupervised domain adaptation has been explored to leverage the knowledge from a labeled source domain to an unlabeled target domain, existing methods focus on the distribution alignment between two domains. However, how to better align source and target features is not well addressed. In this paper, we propose a deep feature registration (DFR) model to generate registered features that maintain domain invariant features and simultaneously minimize the domain-dissimilarity of registered features and target features via histogram matching. We further employ a pseudo label refinement process, which considers both probabilistic soft selection and center-based hard selection to improve the quality of pseudo labels in the target domain. Extensive experiments on multiple UDA benchmarks demonstrate the effectiveness of our DFR model, resulting in new state-of-the-art performance.

cs.CV↗

arXiv2023

LaksNet: an end-to-end deep learning model for self-driving cars in Udacity simulator

The majority of road accidents occur because of human errors, including distraction, recklessness, and drunken driving. One of the effective ways to overcome this dangerous situation is by implementing self-driving technologies in vehicles. In this paper, we focus on building an efficient deep-learning model for self-driving cars. We propose a new and effective convolutional neural network model called `LaksNet' consisting of four convolutional layers and two fully connected layers. We conduct extensive experiments using our LaksNet model with the training data generated from the Udacity simulator. Our model outperforms many existing pre-trained ImageNet and NVIDIA models in terms of the duration of the car for which it drives without going off the track on the simulator.

cs.CV↗

arXiv2023

Complex Image Generation SwinTransformer Network for Audio Denoising

Achieving high-performance audio denoising is still a challenging task in real-world applications. Existing time-frequency methods often ignore the quality of generated frequency domain images. This paper converts the audio denoising problem into an image generation task. We first develop a complex image generation SwinTransformer network to capture more information from the complex Fourier domain. We then impose structure similarity and detailed loss functions to generate high-quality images and develop an SDR loss to minimize the difference between denoised and clean audios. Extensive experiments on two benchmark datasets demonstrate that our proposed model is better than state-of-the-art methods.

cs.SD↗

arXiv2023

Stall Number Detection of Cow Teats Key Frames

In this paper, we present a small cow stall number dataset named CowStallNumbers, which is extracted from cow teat videos with the goal of advancing cow stall number detection. This dataset contains 1042 training images and 261 test images with the stall number ranging from 0 to 60. In addition, we fine-tuned a ResNet34 model and augmented the dataset with the random crop, center crop, and random rotation. The experimental result achieves a 92% accuracy in stall number recognition and a 40.1% IoU score in stall number position prediction.

cs.CV↗

arXiv2023

Lung segmentation with NASNet-Large-Decoder Net

Lung cancer has emerged as a severe disease that threatens human life and health. The precise segmentation of lung regions is a crucial prerequisite for localizing tumors, which can provide accurate information for lung image analysis. In this work, we first propose a lung image segmentation model using the NASNet-Large as an encoder and then followed by a decoder architecture, which is one of the most commonly used architectures in deep learning for image segmentation. The proposed NASNet-Large-decoder architecture can extract high-level information and expand the feature map to recover the segmentation map. To further improve the segmentation results, we propose a post-processing layer to remove the irrelevant portion of the segmentation map. Experimental results show that an accurate segmentation model with 0.92 dice scores outperforms state-of-the-art performance.

eess.IV↗

arXiv2022

BirdSoundsDenoising: Deep Visual Audio Denoising for Bird Sounds

Audio denoising has been explored for decades using both traditional and deep learning-based methods. However, these methods are still limited to either manually added artificial noise or lower denoised audio quality. To overcome these challenges, we collect a large-scale natural noise bird sound dataset. We are the first to transfer the audio denoising problem into an image segmentation problem and propose a deep visual audio denoising (DVAD) model. With a total of 14,120 audio images, we develop an audio ImageMask tool and propose to use a few-shot generalization strategy to label these images. Extensive experimental results demonstrate that the proposed model achieves state-of-the-art performance. We also show that our method can be easily generalized to speech denoising, audio separation, audio enhancement, and noise estimation.

cs.SD↗

arXiv2022

House Price Prediction Based On Deep Learning

Since ancient times, what Chinese people have been pursuing is very simple, which is nothing more than "to live and work happily, to eat and dress comfortable". Today, more than 40 years after the reform and opening, people have basically solved the problem of food and clothing, and the urgent problem is housing. Nowadays, due to the storm of long-term rental apartment intermediary platforms such as eggshell, increasing the sense of insecurity of renters, as well as the urbanization in recent years and the scramble for people in major cities, this will make the future real estate market competition more intense. In order to better grasp the real estate price, let consumers buy a house reasonably, and provide a reference for the government to formulate policies, this paper summarizes the existing methods of house price prediction and proposes a house price prediction method based on mixed depth vision and text features.

cs.LG↗

arXiv2021

A Survey of Unsupervised Domain Adaptation for Visual Recognition

While huge volumes of unlabeled data are generated and made available in many domains, the demand for automated understanding of visual data is higher than ever before. Most existing machine learning models typically rely on massive amounts of labeled training data to achieve high performance. Unfortunately, such a requirement cannot be met in real-world applications. The number of labels is limited and manually annotating data is expensive and time-consuming. It is often necessary to transfer knowledge from an existing labeled domain to a new domain. However, model performance degrades because of the differences between domains (domain shift or dataset bias). To overcome the burden of annotation, Domain Adaptation (DA) aims to mitigate the domain shift problem when transferring knowledge from one domain into another similar but different domain. Unsupervised DA (UDA) deals with a labeled source domain and an unlabeled target domain. The principal objective of UDA is to reduce the domain discrepancy between the labeled source data and unlabeled target data and to learn domain-invariant representations across the two domains during training. In this paper, we first define UDA problem. Secondly, we overview the state-of-the-art methods for different categories of UDA from both traditional methods and deep learning based methods. Finally, we collect frequently used benchmark datasets and report results of the state-of-the-art methods of UDA on visual recognition problem.

cs.CV↗

arXiv2021

Deep Least Squares Alignment for Unsupervised Domain Adaptation

Unsupervised domain adaptation leverages rich information from a labeled source domain to model an unlabeled target domain. Existing methods attempt to align the cross-domain distributions. However, the statistical representations of the alignment of the two domains are not well addressed. In this paper, we propose deep least squares alignment (DLSA) to estimate the distribution of the two domains in a latent space by parameterizing a linear model. We further develop marginal and conditional adaptation loss to reduce the domain discrepancy by minimizing the angle between fitting lines and intercept differences and further learning domain invariant features. Extensive experiments demonstrate that the proposed DLSA model is effective in aligning domain distributions and outperforms state-of-the-art methods.

cs.LG↗

arXiv2021

Enhanced Separable Disentanglement for Unsupervised Domain Adaptation

Domain adaptation aims to mitigate the domain gap when transferring knowledge from an existing labeled domain to a new domain. However, existing disentanglement-based methods do not fully consider separation between domain-invariant and domain-specific features, which means the domain-invariant features are not discriminative. The reconstructed features are also not sufficiently used during training. In this paper, we propose a novel enhanced separable disentanglement (ESD) model. We first employ a disentangler to distill domain-invariant and domain-specific features. Then, we apply feature separation enhancement processes to minimize contamination between domain-invariant and domain-specific features. Finally, our model reconstructs complete feature vectors, which are used for further disentanglement during the training phase. Extensive experiments from three benchmark datasets outperform state-of-the-art methods, especially on challenging cross-domain tasks.

cs.CV↗

arXiv2021

Automatic Head Overcoat Thickness Measure with NASNet-Large-Decoder Net

Transmission electron microscopy (TEM) is one of the primary tools to show microstructural characterization of materials as well as film thickness. However, manual determination of film thickness from TEM images is time-consuming as well as subjective, especially when the films in question are very thin and the need for measurement precision is very high. Such is the case for head overcoat (HOC) thickness measurements in the magnetic hard disk drive industry. It is therefore necessary to develop software to automatically measure HOC thickness. In this paper, for the first time, we propose a HOC layer segmentation method using NASNet-Large as an encoder and then followed by a decoder architecture, which is one of the most commonly used architectures in deep learning for image segmentation. To further improve segmentation results, we are the first to propose a post-processing layer to remove irrelevant portions in the segmentation result. To measure the thickness of the segmented HOC layer, we propose a regressive convolutional neural network (RCNN) model as well as orthogonal thickness calculation methods. Experimental results demonstrate a higher dice score for our model which has lower mean squared error and outperforms current state-of-the-art manual measurement.

cs.CV↗

arXiv2021

Correlated Adversarial Joint Discrepancy Adaptation Network

Domain adaptation aims to mitigate the domain shift problem when transferring knowledge from one domain into another similar but different domain. However, most existing works rely on extracting marginal features without considering class labels. Moreover, some methods name their model as so-called unsupervised domain adaptation while tuning the parameters using the target domain label. To address these issues, we propose a novel approach called correlated adversarial joint discrepancy adaptation network (CAJNet), which minimizes the joint discrepancy of two domains and achieves competitive performance with tuning parameters using the correlated label. By training the joint features, we can align the marginal and conditional distributions between the two domains. In addition, we introduce a probability-based top-$\mathcal{K}$ correlated label ($\mathcal{K}$-label), which is a powerful indicator of the target domain and effective metric to tune parameters to aid predictions. Extensive experiments on benchmark datasets demonstrate significant improvements in classification accuracy over the state of the art.

cs.CV↗

arXiv2021

Deep Spherical Manifold Gaussian Kernel for Unsupervised Domain Adaptation

Unsupervised Domain adaptation is an effective method in addressing the domain shift issue when transferring knowledge from an existing richly labeled domain to a new domain. Existing manifold-based methods either are based on traditional models or largely rely on Grassmannian manifold via minimizing differences of single covariance matrices of two domains. In addition, existing pseudo-labeling algorithms inadequately consider the quality of pseudo labels in aligning the conditional distribution between two domains. In this work, a deep spherical manifold Gaussian kernel (DSGK) framework is proposed to map the source and target subspaces into a spherical manifold and reduce the discrepancy between them by embedding both extracted features and a Gaussian kernel. To align the conditional distributions, we further develop an easy-to-hard pseudo label refinement process to improve the quality of the pseudo labels and then reduce categorical spherical manifold Gaussian kernel geodesic loss. Extensive experimental results show that DSGK outperforms state-of-the-art methods, especially on challenging cross-domain learning tasks.

cs.CV↗

arXiv2021

Efficient Pre-trained Features and Recurrent Pseudo-Labeling in Unsupervised Domain Adaptation

Domain adaptation (DA) mitigates the domain shift problem when transferring knowledge from one annotated domain to another similar but different unlabeled domain. However, existing models often utilize one of the ImageNet models as the backbone without exploring others, and fine-tuning or retraining the backbone ImageNet model is also time-consuming. Moreover, pseudo-labeling has been used to improve the performance in the target domain, while how to generate confident pseudo labels and explicitly align domain distributions has not been well addressed. In this paper, we show how to efficiently opt for the best pre-trained features from seventeen well-known ImageNet models in unsupervised DA problems. In addition, we propose a recurrent pseudo-labeling model using the best pre-trained features (termed PRPL) to improve classification performance. To show the effectiveness of PRPL, we evaluate it on three benchmark datasets, Office+Caltech-10, Office-31, and Office-Home. Extensive experiments show that our model reduces computation time and boosts the mean accuracy to 98.1%, 92.4%, and 81.2%, respectively, substantially outperforming the state of the art.

cs.CV↗

arXiv2021

Adversarial Regression Learning for Bone Age Estimation

Estimation of bone age from hand radiographs is essential to determine skeletal age in diagnosing endocrine disorders and depicting the growth status of children. However, existing automatic methods only apply their models to test images without considering the discrepancy between training samples and test samples, which will lead to a lower generalization ability. In this paper, we propose an adversarial regression learning network (ARLNet) for bone age estimation. Specifically, we first extract bone features from a fine-tuned Inception V3 neural network and propose regression percentage loss for training. To reduce the discrepancy between training and test data, we then propose adversarial regression loss and feature reconstruction loss to guarantee the transition from training data to test data and vice versa, preserving invariant features from both training and test data. Experimental results show that the proposed model outperforms state-of-the-art methods.

cs.CV↗