Global Semantic Information Research Articles

A Semantic Information Decomposition Network for Accurate Segmentation of Texture Defects

Defect detection on textured surfaces remains a challenging task in the field of industrial automation due to the wide range of textures and defects. Current unsupervised learning-based texture defect detection methods based on texture background reconstruction, cannot detect texture defects with high precision because it is difficult to guarantee a high-precision reconstruction of the texture background while suppressing the defect foreground. In this study, we propose a novel semantic information decomposition network (SIDN) for accurate texture defect segmentation. The SIDN is trained on artificial defective images produced by a defect generation module (DGM). First, the SIDN uses a feature extraction module (FEM) to extract latent features with both texture semantic information and defect semantic information. Then, a novel feature separation extraction module (FSEM) for decomposing the texture semantic information and defect semantic information from the feature map generated by the FEM is proposed, preventing the coupling of the texture and defect semantic information from affecting the final segmentation accuracy. Next, a novel global semantic relation module (GSRM) is proposed to determine the relevance of the global semantic information to comprehensively consider the context and improve the feature representation. Finally, a segmentation module (SM) that directly segments the textures and defects instead of reconstructing the texture background is proposed. The final detection result is obtained by calculating a weighted average of the texture and defect segmentation results. The extensive experimental tests with the most popular and most challenging texture defect dataset demonstrate that the SIDN achieves accurate segmentation of various texture defects without using real defect samples.

Simulated Quantum Mechanics-Based Joint Learning Network for Stroke Lesion Segmentation and TICI Grading.

Segmenting stroke lesions and assessing the thrombolysis in cerebral infarction (TICI) grade are two important but challenging prerequisites for an auxiliary diagnosis of the stroke. However, most previous studies have focused only on a single one of two tasks, without considering the relation between them. In our study, we propose a simulated quantum mechanics-based joint learning network (SQMLP-net) that simultaneously segments a stroke lesion and assesses the TICI grade. The correlation and heterogeneity between the two tasks are tackled with a single-input double-output hybrid network. SQMLP-net has a segmentation branch and a classification branch. These two branches share an encoder, which extracts and shares the spatial and global semantic information for the segmentation and classification tasks. Both tasks are optimized by a novel joint loss function that learns the intra- and inter-task weights between these two tasks. Finally, we evaluate SQMLP-net with a public stroke dataset (ATLAS R2.0). SQMLP-net obtains state-of-the-art metrics (Dice:70.98% and accuracy:86.78%) and outperforms single-task and existing advanced methods. An analysis found a negative correlation between the severity of TICI grading and the accuracy of stroke lesion segmentation.

TransFusion-net for multifocus microscopic biomedical image fusion

Background and objectiveDue to the depth of focus (DOF) limitations of the optical systems of microscopes, it is often difficult to achieve full clarity from microscopic biomedical images under high-magnification microscopy. Multifocus microscopic biomedical image fusion (MFBIF) can effectively solve this problem. Considering both information richness and visual authenticity, this paper proposes a transformer network for MFBIF called TransFusion-Net. MethodsTransFusion-Net consists of two modules. One module is an interlayer cross-attention module, which is used to obtain feature mappings under the long-range dependencies observed among multiple nonfocus source images. The other module is a spatial attention upsampling network (SAU-Net) module, which is used to obtain global semantic information after further spatial attention is applied. Thus, TransFusion-Net can simultaneously receive multiple input images from a nonfull-focus microscope and make full use of the strong correlations between the source images to output accurate fusion results in an end-to-end manner. ResultsThe fusion results were quantitatively and qualitatively compared with those of eight state-of-the-art algorithms. In the quantitative experiments, five evaluation metrics, QAB/F, QMI, QAVG, QCB, and PSNR, were used to evaluate the performance of each method, and the proposed method achieved values of 0.6574, 8.4572, 5.6305, 0.7341, and 89.5685, respectively, which are higher than those of the current state-of-the-art algorithms. In the qualitative experiments, a differential image was used for further validation, and the near-zero residuals visually verified the adequacy of the proposed method for fusion. Furthermore, we showed some fusion results of multifocused biomedical microscopy images to verify the reliability of the proposed method, which shows high-quality fusion results. ConclusionMultifocus biomedical microscopic image fusion can be accurately and effectively achieved by devising a deep convolutional neural network with joint cross-attention and spatial attention mechanisms.

TranSEFusionNet: Deep fusion network for colorectal polyp segmentation

Fully convolutional neural (FCN) networks like U-Net have been the state-of-the-art methods in colorectal polyp segmentation. However, U-Net still has some limitations in modelling remote semantic information, especially since the semantic information at different levels can vary greatly, making it difficult to utilize this information fully. To address these issues, we propose a new network architecture called TranSEFusionNet that utilizes the Transformer’s global modelling capability to better focus on global contextual semantic information. In addition, we added two feature fusion modules, a spatial feature fusion module (SFM) and an edge feature fusion module (EFM), to the network. SEF with a skip connection can improve the accuracy of passing deep features to shallow features. EFM in the output part of each decoder layer improves the recognition of edge ambiguous features by refining the semantic information of the network. We validate the model’s performance on five publicly available colorectal polyp datasets, and the experiments show that TranSEFusionNet has higher segmentation accuracy. To measure the generalization ability of TranSEFusionNet, we further applied the model to the cell nuclei dataset, which further verifies the performance of our model.Code: https://github.com/Linaaalin/TranSEFusionNet.

Trainable guided attention based robust leather defect detection

Automatic leather defect detection is becoming increasingly important as a crucial requirement for industry 4.0. It is a highly challenging problem due to the varying appearance and characteristics of various defect types. The diversified scale and spatial position of the defects, and high intra-class variance in addition to inter-class similarity, make the problem highly challenging. A novel optimized backbone network that uses reduced filters to extract global semantic information for defect detection. The feature pyramid network capture defects of different sizes, the spatial attention focuses the detector on precise defect locations, whereas the cross-channel guidance along with channel attention aid the network to discriminate between defects of similar and different classes. The current state bounding box regression methods are employed in conjunction with the proposed OAFE module for precise object localization.

A Multiscale Local–Global Feature Fusion Method for SAR Image Classification with Bayesian Hyperparameter Optimization Algorithm

In recent years, the advancement of deep learning technology has led to excellent performance in synthetic aperture radar (SAR) automatic target recognition (ATR) technology. However, due to the interference of speckle noise, the task of classifying SAR images remains challenging. To address this issue, a multi-scale local–global feature fusion network (MFN) integrating a convolution neural network (CNN) and a transformer network was proposed in this study. The proposed network comprises three branches: a CovNeXt-SimAM branch, a Swin Transformer branch, and a multi-scale feature fusion branch. The CovNeXt-SimAM branch extracts local texture detail features of the SAR images at different scales. By incorporating the SimAM attention mechanism to the CNN block, the feature extraction capability of the model was enhanced from the perspective of spatial and channel attention. Additionally, the Swin Transformer branch was employed to extract SAR image global semantic information at different scales. Finally, the multi-scale feature fusion branch was used to fuse local features and global semantic information. Moreover, to overcome the problem of poor accuracy and inefficiency of the model due to empirically determined model hyperparameters, the Bayesian hyperparameter optimization algorithm was used to determine the optimal model hyperparameters. The model proposed in this study achieved average recognition accuracies of 99.26% and 94.27% for SAR vehicle targets under standard operating conditions (SOCs) and extended operating conditions (EOCs), respectively, on the MSTAR dataset. Compared with the baseline model, the recognition accuracy has been improved by 12.74% and 25.26%, respectively. The results demonstrated that Bayes-MFN reduces the inter-class distance of the SAR images, resulting in more compact classification features and less interference from speckle noise. Compared with other mainstream models, the Bayes-MFN model exhibited the best classification performance.

CFormerFaceNet: Efficient Lightweight Network Merging a CNN and Transformer for Face Recognition

Most face recognition methods rely on deep convolutional neural networks (CNNs) that construct multiple layers of processing units in a cascaded form and employ convolution operations to fuse local features. However, these methods are not conducive to modeling the global semantic information of the face and lack attention to important facial feature regions and their spatial relationships. In this work, a Group Depth-Wise Transpose Attention (GDTA) block is designed to effectively capture both local and global representations, mitigate the issue of limited receptive fields in CNNs, and establish long-range dependencies among different feature regions. Based on GDTA and CNNs, a novel, efficient, and lightweight face recognition model called CFormerFaceNet, which combines a CNN and Transformer, is proposed. The model significantly reduces the parameters and computational cost without compromising performance, greatly improving the computational efficiency of deep neural networks in face recognition tasks. The model achieves competitive accuracy on multiple challenging benchmark face datasets, including LFW, CPLFW, CALFW, SLLFW, CFP_FF, CFP_FP, and AgeDB-30, while maintaining the minimum computational cost compared to all other advanced face recognition models. The experimental results using computers and embedded devices also demonstrate that it can meet real-time requirements in practical applications.

Uniting Multi-Scale Local Feature Awareness and the Self-Attention Mechanism for Named Entity Recognition

In recent years, a huge amount of text information requires processing to support the diagnosis and treatment of diabetes in the medical field; therefore, the named entity recognition of diabetes (DNER) is giving rise to the popularity of this research topic within this particular field. Although the mainstream methods for Chinese medical named entity recognition can effectively capture global context information, they ignore the potential local information in sentences, and hence cannot extract the local context features through an efficient framework. To overcome these challenges, this paper constructs a diabetes corpus and proposes the RMBC (RoBERTa Multi-scale CNN BiGRU Self-attention CRF) model. This model is a named entity recognition model that unites multi-scale local feature awareness and the self-attention mechanism. This paper first utilizes RoBERTa-wwm to encode the characters; then, it designs a local context-wise module, which captures the context information containing locally important features by fusing multi-window attention with residual convolution at the multi-scale and adds a self-attention mechanism to address the restriction of the bidirectional gated recurrent unit (BiGRU) capturing long-distance dependencies and to obtain global semantic information. Finally, conditional random fields (CRF) are relied on to learn of the dependency between adjacent tags and to obtain the optimal tag sequence. The experimental results on our constructed private dataset, termed DNER, along with two benchmark datasets, demonstrate the effectiveness of the model in this paper.

HTC-Net: Hashimoto's thyroiditis ultrasound image classification model based on residual network reinforced by channel attention mechanism.

Convolutional neural network (CNN) is efficient in extracting and aggregating local features in the spatial dimension of the images. However, obtaining the inapparent texture information of the low-echo area in the ultrasound images is not easy, and it is especially challenging for the early lesion recognition in Hashimoto's thyroiditis (HT) ultrasound images. In this paper, a HT ultrasound image classification model HTC-Net based on residual network reinforced by channel attention mechanism is proposed. HTC-Net strengthens the features of the important channels by reinforced channel attention mechanism through which the high-level semantic information is enchanced and the low-level semantic information is suppressed. Residual network assists HTC-Net focus on the key local areas of the ultrasound images while pay attention to the global semantic information. Furthermore, in order to solve the problem of uneven distribution caused by large amount of difficult-to-classify samples in the data sets, a new feature loss function TanCELoss with weight factor dynamically adjusting is constructed. TanCELoss function can better assist HTC-Net to transform difficult-to-classify samples into easy-to-classify samples gradually, and improve the balancing distribution of the samples. The experiments are implemented based on data sets collected by the Endocrinology Department of four branches from Guangdong Provincial Hospital of Chinese Medicine. Both quantitative testing and visualization results show that HTC-Net obtains STOA performance for early lesions recognition in HT ultrasound images. HTC-Net has great application value especially under the condition of owning only small data samples.

HIGF-Net: Hierarchical information-guided fusion network for polyp segmentation based on transformer and convolution feature learning

Polyp segmentation plays a role in image analysis during colonoscopy screening, thus improving the diagnostic efficiency of early colorectal cancer. However, due to the variable shape and size characteristics of polyps, small difference between lesion area and background, and interference of image acquisition conditions, existing segmentation methods have the phenomenon of missing polyp and rough boundary division. To overcome the above challenges, we propose a multi-level fusion network called HIGF-Net, which uses hierarchical guidance strategy to aggregate rich information to produce reliable segmentation results. Specifically, our HIGF-Net excavates deep global semantic information and shallow local spatial features of images together with Transformer encoder and CNN encoder. Then, Double-stream structure is used to transmit polyp shape properties between feature layers at different depths. The module calibrates the position and shape of polyps in different sizes to improve the model's efficient use of the rich polyp features. In addition, Separate Refinement module refines the polyp profile in the uncertain region to highlight the difference between the polyp and the background. Finally, in order to adapt to diverse collection environments, Hierarchical Pyramid Fusion module merges the features of multiple layers with different representational capabilities. We evaluate the learning and generalization abilities of HIGF-Net on five datasets using six evaluation metrics, including Kvasir-SEG, CVC-ClinicDB, ETIS, CVC-300, and CVC-ColonDB. Experimental results show that the proposed model is effective in polyp feature mining and lesion identification, and its segmentation performance is better than ten excellent models.

Dual-branch vision transformer for blind image quality assessment

Blind image quality assessment (BIQA) has always been a challenging problem due to the absence of reference images. In this paper, we propose a novel dual-branch vision transformer for BIQA, which simultaneously considers both local distortions and global semantic information. It first extracts dual-scale features from the backbone network, and then each scale feature is fed into one of the transformer encoder branches as a local feature embedding to consider the scale-variant local distortions. Each transformer branch obtains the context of global image distortion as well as the local distortion by adopting content-aware embedding. Finally, the outputs of the dual-branch vision transformer are combined by using multiple feed-forward blocks to predict the image quality scores effectively. Experimental results demonstrate that the proposed BIQA method outperforms the conventional methods on the six public BIQA datasets.

Focal cross transformer: multi-view brain tumor segmentation model based on cross window and focal self-attention.

Recently, the Transformer model and its variants have been a great success in terms of computer vision, and have surpassed the performance of convolutional neural networks (CNN). The key to the success of Transformer vision is the acquisition of short-term and long-term visual dependencies through self-attention mechanisms; this technology can efficiently learn global and remote semantic information interactions. However, there are certain challenges associated with the use of Transformers. The computational cost of the global self-attention mechanism increases quadratically, thus hindering the application of Transformers for high-resolution images. In view of this, this paper proposes a multi-view brain tumor segmentation model based on cross windows and focal self-attention which represents a novel mechanism to enlarge the receptive field by parallel cross windows and improve global dependence by using local fine-grained and global coarse-grained interactions. First, the receiving field is increased by parallelizing the self-attention of horizontal and vertical fringes in the cross window, thus achieving strong modeling capability while limiting the computational cost. Second, the focus on self-attention with regards to local fine-grained and global coarse-grained interactions enables the model to capture short-term and long-term visual dependencies in an efficient manner. Finally, the performance of the model on Brats2021 verification set is as follows: dice Similarity Score of 87.28, 87.35 and 93.28%; Hausdorff Distance (95%) of 4.58 mm, 5.26 mm, 3.78 mm for the enhancing tumor, tumor core and whole tumor, respectively. In summary, the model proposed in this paper has achieved excellent performance while limiting the computational cost.

Cerebrovascular segmentation from mesoscopic optical images using Swin Transformer

Vascular segmentation is a crucial task in biomedical image processing, which is significant for analyzing and modeling vascular networks under physiological and pathological states. With advances in fluorescent labeling and mesoscopic optical techniques, it has become possible to map the whole-mouse-brain vascular networks at capillary resolution. However, segmenting vessels from mesoscopic optical images is a challenging task. The problems, such as vascular signal discontinuities, vessel lumens, and background fluorescence signals in mesoscopic optical images, belong to global semantic information during vascular segmentation. Traditional vascular segmentation methods based on convolutional neural networks (CNNs) have been limited by their insufficient receptive fields, making it challenging to capture global semantic information of vessels and resulting in inaccurate segmentation results. Here, we propose SegVesseler, a vascular segmentation method based on Swin Transformer. SegVesseler adopts 3D Swin Transformer blocks to extract global contextual information in 3D images. This approach is able to maintain the connectivity and topology of blood vessels during segmentation. We evaluated the performance of our method on mouse cerebrovascular datasets generated from three different labeling and imaging modalities. The experimental results demonstrate that the segmentation effect of our method is significantly better than traditional CNNs and achieves state-of-the-art performance.

Intelligent Identification Method of Hydrophobic Grade of Composite Insulator Based on Efficient GA‐YOLO Former Network

AbstractThe hydrophobia of composite insulators is a parameter closely related to insulation performance. As an essential index to measure the pollution flashover resistance of composite insulators, the hydrophobic grade is of great significance to maintaining transmission line insulators. To judge the hydrophobic grade of composite insulators accurately and efficiently and to solve the problem of difficult extraction of fine‐grained feature information and high similarity between classes in the data sets of anti‐hydrophobic grade recognition, an end‐to‐end intelligent identification method of the hydrophobic grade of composite insulators based on Efficient GA‐YOLO Former network is proposed. This method introduces Swin Transformer—a multi‐head self‐attention mechanism with shifted windows, into CNN, which helps the model better understand the multi‐scale global semantic information of images through cross‐layer interaction. The efficient BiFPN structure is introduced in the neck part to realize the bidirectional cross‐scale connection and weighted feature fusion in the feature extraction process. By integrating the GAM attention mechanism to amplify the significant cross‐dimensional acceptance region, the model can learn more subtle details that are more difficult to distinguish, improving model efficiency and detection accuracy. Combined with the target detection algorithm, a system for evaluating the hydrophobic grade of composite insulators on transmission lines is designed, which can realize the real‐time measurement of the hydrophobic grade of composite insulators and has practical engineering application value. © 2023 Institute of Electrical Engineer of Japan and Wiley Periodicals LLC.

Multi-scale semantic enhancement network for object detection

In the field of object detection, feature pyramid network (FPN) can effectively extract multi-scale information. However, the majority of FPN-based methods suffer from a semantic gap between features of various sizes before feature fusion, which can lead to feature maps with significant aliasing. In this paper, we present a novel multi-scale semantic enhancement feature pyramid network (MSE-FPN) which consists of three effective modules: semantic enhancement module, semantic injection module, and gated channel guidance module to alleviate these problems. Specifically, inspired by the strong ability of the self-attention mechanism to model context, we propose a semantic enhancement module to model global context to obtain the global semantic information before feature fusion. Then we propose the semantic injection module to divide and merge global semantic information into feature maps at various scales to narrow the semantic gap between features at different scales and efficiently utilize the semantic information of high-level features. Finally, to mitigate feature aliasing caused by feature fusion, the gated channel guidance module selectively outputs crucial features via a gating unit. By replacing FPN with MSE-FPN in Faster R-CNN, our models achieve 39.4 and 41.2 Average precision (AP) using ResNet50 and ResNet101 as the backbone network respectively. When using ResNet-101-64x4d as the backbone, MSE-FPN achieved up to 43.4 AP. Our results demonstrate that replacing FPN with MSE-FPN significantly enhances the detection performance of state-of-the-art FPN-based detectors.

An efficient change detection method for disaster-affected buildings based on a lightweight residual block in high-resolution remote sensing images

ABSTRACT After a natural disaster, it is critical to perform a damage assessment of affected buildings rapidly and accurately to meet emergency response requirements. With the development of deep learning (DL) technology, the change detection (CD) method based on fully convolutional networks has been used for disaster analysis and assessment. However, the main limitations of most present methods are their large number of parameters, high computational complexity, and limited deployment equipment resources, which affect the real-time performance of disaster analysis. Therefore, to make a trade-off in the relationship between the accuracy and efficiency of damaged building CD, this research proposes an efficient method based on a lightweight residual block (LRB) in bitemporal high-resolution remote sensing (RS) images (LRBNet). LRBNet is an encoder – decoder structure with the Siamese network and UNet++ as the backbone. It uses our LRB in the whole network to replace the original convolution unit, which can extract features to reduce computational complexity. The LRB consists of the lightweight compression module (LCM) and efficient channel attention (ECA) to learn the essential damage information in the feature map and improve the feature expression ability efficiently. Moreover, the multilevel damage feature aggregation attention module is the last process of the network and performs fine-grained aggregation of the multilevel disaster change features and emphasizes global semantic and spatial information to maximize the representative damage change characteristics. Furthermore, we conducted a series of experiments on the global-scale disaster xBD dataset (including 19 different disaster events). Our method not only accurately evaluated the level of disaster but also had favourable detection efficiency, achieving a good trade-off between recognition accuracy and computational resources.

Sequence labeling with MLTA: Multi-level topic-aware mechanism

Current sequence labeling systems are typically based on CNNs-BiLSTM-CRF, which has been viewed as a standard method. However, on one hand, the standard method often uses static word embedding to embed input words without considering the polysemy. On the other hand, BiLSTM has limitations in modeling the global context information, which further makes the standard method hard to correctly recognize the label of words that are ambiguous. In this paper, we propose a multi-level topic-aware mechanism to obtain word-level and corpus-level topic representations for sequence labeling tasks. The word-level topic representation can capture different word sense to some extent and enhance the discriminative of each word. The corpus-level topic representation is a kind of global semantic information which is helpful for the understanding of a word in different contexts. Both of the word-level and corpus-level topic representations are adaptively fused with sequential information extracted by BiLSTM via a proposed fusion gate. Experimental results validate the superior performance of the proposed model compared with state-of-the-art variants of the standard methods. In particular, the multi-level topic-aware mechanism is helpful for achieving better performances on words that are ambiguous.

Hybrid semantic segmentation for tunnel lining cracks based on Swin Transformer and convolutional neural network

AbstractIn the field of tunnel lining crack identification, the semantic segmentation algorithms based on convolution neural network (CNN) are extensively used. Owing to the inherent locality of CNN, these algorithms cannot make full use of context semantic information, resulting in difficulty in capturing the global features of crack. Transformer‐based networks can capture global semantic information, but this method also has the deficiencies of strong data dependence and easy loss of local features. In this paper, a hybrid semantic segmentation algorithm for tunnel lining crack, named SCDeepLab, is proposed by fusing Swin Transformer and CNN in the encoding and decoding framework of DeepLabv3+ to address the above issues. In SCDeepLab, a joint backbone network is introduced with CNN‐based Inverse Residual Block and Swin Transformer Block. The former is used to extract the local detailed information of the crack to generate the shallow feature layer, whereas the latter is used to extract the global semantic information to obtain the deep feature layer. In addition, Efficient Channel Attention enhanced Feature Fusion Module is proposed to fuse the shallow and deep features to combine the advantages of the two types of features. Furthermore, the strategy of transfer learning is adopted to solve the data dependency of Swin Transformer. The results show that the mean intersection over union (mIoU) and mean pixel accuracy (mPA) of SCDeepLab on the data sets constructed in this paper are 77.41% and 84.42%, respectively, which have higher segmentation accuracy than previous CNN‐based and transformer‐based semantic segmentation algorithms.

GL-Segnet: Global-Local representation learning net for medical image segmentation.

Medical image segmentation has long been a compelling and fundamental problem in the realm of neuroscience. This is an extremely challenging task due to the intensely interfering irrelevant background information to segment the target. State-of-the-art methods fail to consider simultaneously addressing both long-range and short-range dependencies, and commonly emphasize the semantic information characterization capability while ignoring the geometric detail information implied in the shallow feature maps resulting in the dropping of crucial features. To tackle the above problem, we propose a Global-Local representation learning net for medical image segmentation, namely GL-Segnet. In the Feature encoder, we utilize the Multi-Scale Convolution (MSC) and Multi-Scale Pooling (MSP) modules to encode the global semantic representation information at the shallow level of the network, and multi-scale feature fusion operations are applied to enrich local geometric detail information in a cross-level manner. Beyond that, we adopt a global semantic feature extraction module to perform filtering of irrelevant background information. In Attention-enhancing Decoder, we use the Attention-based feature decoding module to refine the multi-scale fused feature information, which provides effective cues for attention decoding. We exploit the structural similarity between images and the edge gradient information to propose a hybrid loss to improve the segmentation accuracy of the model. Extensive experiments on medical image segmentation from Glas, ISIC, Brain Tumors and SIIM-ACR demonstrated that our GL-Segnet is superior to existing state-of-art methods in subjective visual performance and objective evaluation.

Trimap-guided feature mining and fusion network for natural image matting

Mining interesting objects with trimap guidance and fusing multi-level features are two important issues for trimap-based matting. For feature mining, existing methods simply feed trimaps to an encoder network and mine features evenly in different regions with a decoding module, which lacks an efficient and effective way to mine interesting objects pointed out by trimaps. For emerging content-based feature fusion, most existing matting methods only focus on local features which lack the guidance of a global feature with strong semantic information related to the interesting object. In this paper, we propose a trimap-guided feature mining and fusion network consisting of our trimap-guided non-background multi-scale pooling (TMP) module and global-local context-aware fusion (GLF) modules. Considering that trimap provides strong semantic guidance, our TMP module focuses effective feature mining on interesting objects under the guidance of trimap without extra parameters. Furthermore, our GLF modules use global semantic information of interesting objects mined by our TMP module to guide an effective global-local context-aware multi-level feature fusion. In addition, we build a common interesting object matting (CIOM) dataset to advance high-quality image matting. Experimental results on the Composition-1k test set, Alphamatting benchmark, and our CIOM test set demonstrate that our method outperforms state-of-the-art approaches. Codes and models will be publicly available at https://github.com/Serge-weihao/TMF-Matting.