Search

Occlusion or posture change of the face in natural scenes has typical asymmetry; however, an asymmetric face plays a key part in the lack of information available for facial expression recognition. To solve the problem of low accuracy of asymmetric facial expression recognition, this paper proposes a fusion of channel global features and a spatial local information expression recognition network called the "Channel–Spatial Integration Network" (CSINet). First, to extract the underlying detail information and deepen the network, the attention residual module with a redundant information filtering function is designed, and the backbone feature-extraction network is constituted by module stacking. Second, considering the loss of information in the local key area of face occlusion, the channel–spatial fusion structure is constructed, and the channel features and spatial features are combined to enhance the accuracy of occluded facial recognition. Finally, before the full connection layer, more local spatial information is embedded into the global channel information to capture the relationship between different channel–spatial targets, which improves the accuracy of feature expression. Experimental results on the natural scene facial expression data sets RAF-DB and FERPlus show that the recognition accuracies of the modeling approach proposed in this paper are 89.67% and 90.83%, which are 13.24% and 11.52% higher than that of the baseline network ResNet50, respectively. Compared with the latest facial expression recognition methods such as CVT, PACVT, etc., the method in this paper obtains better evaluation results of masked facial expression recognition, which provides certain theoretical and technical references for daily facial emotion analysis and human–computer interaction applications. [ABSTRACT FROM AUTHOR]

With the rapid development of computer vision and artificial intelligence technology, visual object detection has made unprecedented progress, and small object detection in complex scenes has attracted more and more attention. To solve the problems of ambiguity, overlap and occlusion in small object detection in complex scenes. In this paper, a multi-scale fusion feature enhanced path aggregation network MSFE-PANet is proposed. By adding attention mechanism and feature fusion, the fusion of strong positioning information of deep feature map and strong semantic information of shallow feature map is enhanced, which helps the network to find interesting areas in complex scenes and improve its sensitivity to small objects. The rejection loss function and network prediction scale are designed to solve the problems of missing detection and false detection of overlapping and blocking small objects in complex backgrounds. The proposed method achieves an accuracy of 40.7% on the VisDrone2021 dataset and 89.7% on the PASCAL VOC dataset. Comparative analysis with mainstream object detection algorithms proves the superiority of this method in detecting small objects in complex scenes. [ABSTRACT FROM AUTHOR]

In category-level object pose estimation tasks, determining how to mitigate intra-class shape variations and improve pose estimation accuracy for complex objects remains a challenging problem to solve. To address this issue, this paper proposes a new network architecture, LA-Net, to efficiently ascertain object poses from features. Firstly, we extend the 3D graph convolution network architecture by introducing the LS-Layer (Linear Connection Layer), which enables the network to acquire features from different layers and perform multi-scale feature fusion. Secondly, LA-Net employs a novel attention mechanism (PSA) and a Max-Pooling layer to extract local and global geometric information, which enhances the network's ability to perceive object poses. Finally, the proposed LA-Net recovers the rotation information of an object by decoupling the rotation mechanism. The experimental results show that LA-Net can has much better accuracy in object pose estimation compared to the baseline method (HS-Pose). Especially for objects with complex shapes, its performance is 8.2% better for the 10°5 cm metric and 5% better for the 10°2 cm metric. [ABSTRACT FROM AUTHOR]

Session-based recommendation (SBR) is to predict the items that users are likely to click afterward by using their recent click history. Learning item features from existing session data to capture users' current preferences is the main problem to be solved in session-based recommendation domain, and fusing global and local information to learn users' preferences is an effective way to obtain this information more accurately. In this paper, we propose a session-based recommendation with fusion of hypergraph item global and context features (FHGIGC), which learns users' current preferences by fusing item global and contextual features. Specifically, the model first constructs a global hypergraph and a local hypergraph and uses the hypergraph neural network to learn item global features and local features by relevant session information and item contextual information, respectively. Then, the learned features are fused by the attention mechanism to obtain the final item features and session features. Finally, personalized recommendations are generated for users based on the fused features. Experiments were conducted on three datasets of session-based recommendation, and the results demonstrate that the FHGIGC model can improve the accuracy of recommendations. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Computer Engineering & Applications is the property of Beijing Journal of Computer Engineering & Applications Journal Co Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Local feature information plays an important role in image segmentation, however, the referring image segmentation task is dependent on the text expression, so it is impossible to extract local feature information directly from the original reference image. In order to solve this problem, this paper proposes a specific noun-guided local feature extraction deep neural network model（NgLFNet). The NgLFNet model can automatically mine the local feature information of the object to be segmented according to the key nouns in the input text expression. Specifically, the model first obtains key nouns in text through sentence analysis; secondly, extracting corresponding features through text and image encoders, and using the key nouns to obtain local features of high-interest regions through the multihead attention mechanism; then the multi-modal features are gradually fused to learn; Finally, the decoding and correction module uses the obtained local features to perform more detailed corrections on the prediction mask to obtain the final result. The method in this paper is compared with a variety of mainstream referring segmentation methods and the experimental results show that the proposed method improves the accuracy of text-based instance segmentation task. [ABSTRACT FROM AUTHOR]

To improve the accuracy of steel surface defect detection, an improved model of multi-directional optimization based on the YOLOv8 algorithm was proposed in this study. First, we innovate the CSP Bottleneck with the two convolutions (C2F) module in YOLOv8 by introducing deformable convolution (DCN) technology to enhance the learning and expression ability of complex texture and irregular shape defect features. Secondly, the advanced Bidirectional Feature Pyramid Network (BiFPN) structure is adopted to realize the weight distribution learning of input features of different scales in the feature fusion stage, allowing for more effective integration of multi-level feature information. Next, the BiFormer attention mechanism is embedded in the backbone network, allowing the model to adaptively allocate attention based on target features, such as flexibly and efficiently skipping non-critical areas, and focusing on identifying potentially defective parts. Finally, we adjusted the loss function from Complete-Intersection over Union (CIoU) to Wise-IoUv3 (WIoUv3) and used its dynamic non-monotony focusing property to effectively solve the problem of overfitting the low quality target bounding box. The experimental results show that the mean Average Precision (mAP) of the improved model in the task of steel surface defect detection reaches 84.8%, which depicts a significant improvement of 6.9% compared with the original YOLO8 model. The improved model can quickly and accurately locate and classify all kinds of steel surface defects in practical applications and meet the needs of steel defect detection in industrial production. [ABSTRACT FROM AUTHOR]

Logo detection is one of the crucial branches in computer vision due to various real-world applications, such as automatic logo detection and recognition, intelligent transportation, and trademark infringement detection. Compared with traditional handcrafted-feature-based methods, deep learning-based convolutional neural networks (CNNs) can learn both low-level and high-level image features. Recent decades have witnessed the great feature representation capabilities of deep CNNs and their variants, which have been very good at discovering intricate structures in high-dimensional data and are thereby applicable to many domains including logo detection. However, logo detection remains challenging, as existing detection methods cannot solve well the problems of a multiscale and large aspect ratios. In this paper, we tackle these challenges by developing a novel long-range dependence involutional network (LDI-Net). Specifically, we designed a strategy that combines a new operator and a self-attention mechanism via rethinking the intrinsic principle of convolution called long-range dependence involution (LD involution) to alleviate the detection difficulties caused by large aspect ratios. We also introduce a multilevel representation neural architecture search (MRNAS) to detect multiscale logo objects by constructing a novel multipath topology. In addition, we implemented an adaptive RoI pooling module (ARM) to improve detection efficiency by addressing the problem of logo deformation. Comprehensive experiments on four benchmark logo datasets demonstrate the effectiveness and efficiency of the proposed approach. [ABSTRACT FROM AUTHOR]

In order to alleviate the adverse effects resulted from complex scenes for object tracking, such as fast movement, mottled background, interference of similar objects, and occlusion etc., an algorithm using dual-template Siamese network with attention feature fusion, named SiamDT, is proposed in this paper. The main idea include that the original ResNet-50 network is improved to extract deep semantic information and shallow spatial information, which are effectively fused using the attention mechanism to achieve accurate feature representation of objects. In addition, a template branch is added to the traditional Siamese network in which a dynamic template is generated together with the first frame image to solve the problems of template failure and model drift. Experimental results on OTB100 dataset and VOT2018 dataset show that the proposed approach obtains the excellent performance compared with the state-of-the-art tracking algorithms, which verifies the feasibility and effectiveness of the proposed approach. [ABSTRACT FROM AUTHOR]

Copyright of Journal of Chongqing University of Technology (Natural Science) is the property of Chongqing University of Technology and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

The specific task of vehicle re-identification is how to quickly and correctly match the same vehicle in different scenarios. In order to solve the problem of inter-class similarity and environmental interference in vehicle images in complex scenes, one fusion attention method is put forward based on the idea of obtaining the distinguishing features of details—the mechanism for the vehicle re-identification method. First, the vehicle image is preprocessed to restore the image's attributes better. Then, the processed image is sent to ResNet50 to extract the features of the second and third layers, respectively. Then, the feature fusion is carried out through the two-layer attention mechanism for a network model. This model can better focus on local detail features, and global features are constructed and named SDLAU-Reid. In the training process, a data augmentation strategy of random erasure is adopted to improve the robustness. The experimental results show that the mAP and rank-k indicators of the model on VeRi-776 and the VehicleID are better than the results of the existing vehicle re-identification algorithms, which verifies the algorithm's effectiveness. [ABSTRACT FROM AUTHOR]

In recent years, intelligent driving technology based on vehicle–road cooperation has gradually become a research hotspot in the field of intelligent transportation. There are many studies regarding vehicle perception, but fewer studies regarding roadside perception. As sensors are installed at different heights, the roadside object scale varies violently, which burdens the optimization of networks. Moreover, there is a large amount of overlapping and occlusion in complex road environments, which leads to a great challenge of object distinction. To solve the two problems raised above, we propose RD-YOLO. Based on YOLOv5s, we reconstructed the feature fusion layer to increase effective feature extraction and improve the detection capability of small targets. Then, we replaced the original pyramid network with a generalized feature pyramid network (GFPN) to improve the adaptability of the network to different scale features. We also integrated a coordinate attention (CA) mechanism to find attention regions in scenarios with dense objects. Finally, we replaced the original Loss with Focal-EIOU Loss to improve the speed of the bounding box regression and the positioning accuracy of the anchor box. Compared to the YOLOv5s, the RD-YOLO improves the mean average precision (mAP) by 5.5% on the Rope3D dataset and 2.9% on the UA-DETRAC dataset. Meanwhile, by modifying the feature fusion layer, the weight of RD-YOLO is decreased by 55.9% while the detection speed is almost unchanged. Nevertheless, the proposed algorithm is capable of real-time detection at faster than 71.9 frames/s (FPS) and achieves higher accuracy than the previous approaches with a similar FPS. [ABSTRACT FROM AUTHOR]

Depth estimation algorithms are of great significance for autonomous vehicle ranging. In order to solve the problems of large errors, fuzzy image generation and detail loss in depth estimation algorithms, this paper proposes an improved algorithm of the MonoDepth2 algorithm. First, the feature extraction network of the original algorithm is changed to the ResNeXt block to improve the running speed and enrich feature information. Secondly, the Convolutional Block Attention Module attention mechanism is added to the depth estimation network to improve the algorithm accuracy and promote detail detection effect. Finally, the Atrous Spatial Pyramid Pooling network structure is added between the encoder and decoder of the depth estimation network to optimize feature fusion at different scales and extract more detailed features. Compared with the original MonoDepth2 algorithm, the improved algorithm reduces the Absolute Relative Error by 0.4 %, the Squared Relative Error by 6.1 %, the Root Mean Square Error by 15 %, and the accuracy(thr=1.25), increased by 0.4 %, The generated depth map recovers deeper image details and a more detailed target contour. [ABSTRACT FROM AUTHOR]

Small object detection is one of the challenging tasks in computer vision. Most of the existing small object detection models cannot fully extract the characteristics of small objects within an image, due to the small coverage area, low resolution and unclear detailed information of small objects in the image; hence, the effect of these models is not ideal. To solve this problem, a simple and efficient reinforce feature pyramid network R-FPN is proposed for the YOLOv5 algorithm. The learnable weight is introduced to show the importance of different input features, make full use of the useful information of different feature layers and strengthen the extraction of small object features. At the same time, a channel space mixed attention CSMA module is proposed to extract the detailed information of small objects combined with spaces and channels, suppress other useless information and further improve the accuracy of small object detection. The experimental results show that the proposed method improves the average accuracy AP, AP50 and AR100 of the original algorithm by 2.11%, 2.86% and 1.94%, respectively, and the detection effect is better than the existing small object detection algorithms, which proves the effectiveness of the proposed method. [ABSTRACT FROM AUTHOR]

The detection of insulator umbrella disc shedding is very important to the stable operation of a transmission line. In order to accomplish the accurate detection of the insulator umbrella disc shedding in foggy weather, a two-stage detection model combined with a defogging algorithm is proposed. In the dehazing stage of insulator images, solving the problem of real hazy image data is difficult; the foggy images are dehazed by the method of synthetic foggy images training and real foggy images fine-tuning. In the detection stage of umbrella disc shedding, a small object detection algorithm named FA-SSD is proposed to solve the problem of the umbrella disc shedding occupying only a small proportion of an aerial image. On the one hand, the shallow feature information and deep feature information are fused to improve the feature extraction ability of small targets; on the other hand, the attention mechanism is introduced to strengthen the feature extraction network's attention to the details of small targets and improve the model's ability to detect the umbrella disc shedding. The experimental results show that our model can accurately detect the insulator umbrella disc shedding defect in the foggy image; the accuracy of the defect detection is 0.925, and the recall is 0.841. Compared with the original model, it improved by 5.9% and 8.6%, respectively. [ABSTRACT FROM AUTHOR]

Existing deep learning based single image super-resolution (SISR) models usually improve the fitting ability of the model by increasing the number of network layers, but fail to fully extract and reuse features, leading low quality of reconstructed images. To solve this problem, this paper proposed an image super-resolution model based on feature fusion and attention mechanism. This model used residual in residual (RIR) structure in feature extraction module. The feature extraction module of the network consisted of several residual groups. Each residual group consisted of several residual blocks. This module implemented local feature fusion in each residual group and global feature fusion between each group. In addition, this model introduced coordinate attention module into each residual block and spatial attention module into each residual group. It verifies that the model is able to fully extract features and reuse features. The final experimental results show that the model is superior to the existing models in objective evaluation indexes and subjective visual effect. [ABSTRACT FROM AUTHOR]

In this paper, we provide external image features and use the internal attention mechanism to solve the VQA problem given a dataset of textual questions and related images. Most previous models for VQA use a pair of images and questions as input. In addition, the model adopts a question-oriented attention mechanism to extract the features of the entire image and then perform feature fusion. However, the shortcoming of these models is that they cannot effectively eliminate the irrelevant features of the image. In addition, the problem-oriented attention mechanism lacks in the mining of image features, which will bring in redundant image features. In this paper, we propose a VQA model based on adversarial learning and bidirectional attention. We exploit external image features that are not related to the question to form an adversarial mechanism to boost the accuracy of the model. Target detection is performed on the image—that is, the image-oriented attention mechanism. The bidirectional attention mechanism is conducive to promoting model attention and eliminating interference. Experimental results are evaluated on benchmark datasets, and our model performs better than other models based on attention methods. In addition, the qualitative results show the attention maps on the images and leads to predicting correct answers. [ABSTRACT FROM AUTHOR]

Copyright of Chinese Journal of Liquid Crystal & Displays is the property of Chinese Journal of Liquid Crystal & Displays and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use. This abstract may be abridged. No warranty is given about the accuracy of the copy. Users should refer to the original published version of the material for the full abstract. (Copyright applies to all Abstracts.)

Cracks on concrete surfaces are vital factors affecting construction safety. Accurate and efficient crack detection can prevent safety-related accidents. Using drones to photograph cracks on a concrete surface and detect them through computer vision technology has the advantages of accurate target recognition, simple practical operation, and low cost. To solve this problem, an improved CenterNet concrete crack-detection model is proposed. Firstly, a channel-space attention mechanism is added to the original model to enhance the ability of the convolution neural network to pay attention to the image. Secondly, a feature selection module is introduced to scale the feature map in the downsampling stage to a uniform size and combine it in the channel dimension. In the upsampling stage, the feature selection module adaptively selects the combined features and fuses them with the output features of the upsampling. Finally, the target size loss is optimized from a Smooth L1 Loss to IoU Loss to lessen its inability to adapt to targets of different sizes. The experimental results show that the improved CenterNet model reduces the FPS by 123.7 Hz, increases the GPU memory by 62 MB, increases the FLOPs by 3.81 times per second, and increases the AP by 15.4% compared with the original model. The GPU memory occupancy remained stable during the training process and exhibited good real-time performance and robustness. [ABSTRACT FROM AUTHOR]

Existing semantic segmentation networks perform well in accuracy by spending much computation. However, for practical applications, not only high segmentation accuracy but also high inference speed is required. To solve the problem of the difficult balance between accuracy and speed, we propose a new real-time semantic segmentation network (FBRNet). To extract multi-scale semantic information more quickly, we propose a lightly weighted reinforced atrous spatial pyramid pooling module (arASPP) based on the attention mechanism, which can extract richer and more advanced features with less computation than the original ASPP. To eliminate the semantic gap between high- and low-level features, we propose a new feature fusion module (CSFM), in which a shuffling mechanism is introduced to enhance robustness, and a parallel contextual information enhancement module and detail information enhancement module are built to facilitate the information exchange between high- and low-level features, achieving the effect of improving the model feature representation. Finally, we also introduce high-level features, fusing Laplace convolution and spatial attention mechanisms, and design the edge feature reinforcement module (LABRM) to eliminate the noise of low-level features and compensate for the model's segmentation effect target boundary. In the Cityscapes validation set and test set, FBRNet achieves 77.63% and 75.3% mIoU, and 101.9 FPS on a single tesla-T4 GPU, also achieved 72.4% mIoU and 89.8 FPS on the CamVid dataset and 55.2% mIoU and 100.8 FPS on the BDD100K dataset, which is a better balance of accuracy and speed compared with existing networks. The code is available at https://github.com/little5570/FBRNet. [ABSTRACT FROM AUTHOR]

Visual object trackers based on Siamese networks perform well in visual object tracking (VOT); however, degradation of the tracking accuracy occurs when the target has fast motion, large-scale changes, and occlusion. In this study, in order to solve this problem and enhance the inference speed of the tracker, fast and accurate visual tracking with a group convolution and pixel-level correlation based on a Siamese network is proposed. The algorithm incorporates multi-layer feature information on the basis of Siamese networks. We designed a multi-scale feature aggregated channel attention block (MCA) and a global-to-local-information-fused spatial attention block (GSA), which enhance the feature extraction capability of the network. The use of a pixel-level mutual correlation operation in the network to match the search region with the template region refines the bounding box and reduces background interference. Comparing our work with the latest algorithms, the precision and success rates on the UAV123, OTB100, LaSOT, and GOT10K datasets were improved, and our tracker was able to run at 40FPS, with a better performance in complex scenes such as those with occlusion, illumination changes, and fast-motion situations. [ABSTRACT FROM AUTHOR]

Ship surveillance plays an important role in ensuring the safety of maritime transportation and navigation. Due to the influence of factors such as waves and special weather, the existing detection methods still cannot balance the accuracy, speed and the parameters of the model in the changeable and complex marine environment. To solve this problem, this paper proposes an improved real-time method based on YOLOv5, which has few parameters and achieves high detection accuracy with little memory and computation cost. Collaborative Attention (CA) mechanism is added to the network structure, which enables the model to more accurately locate and identify target regions. We also design a Spatial Pyramid Pooling module (SPP) and a weighted pyramid network called Bidirectional Feature Pyramid Network (BiFPN) based on the characteristics of the ships to better fuse feature information. Transformer encoder is introduced to capture long-distance dependencies and preserve global and local features to the greatest extent. Furthermore, the ability of our proposed structure to localize objects at each stage is improved through integrating the output of multiple modules. The experimental results show that, the comprehensive performance of this method is better than the existing technology in ship detection on different evaluation criteria. • Collaborative Attention is used to more accurately locate and identify the target • Spatial Pyramid Pool is put forward to better use the feature information • Weighted feature fusion is introduced according to the characteristics of the ships • Transformer encoder is added to integrate contextual information • The proposed network has been improved in detection accuracy and model complexity [ABSTRACT FROM AUTHOR]

Graph convolutional networks are widely used in skeleton-based action recognition because of their good fitting ability to non-Euclidean data. While conventional multi-scale temporal convolution uses several fixed-size convolution kernels or dilation rates at each layer of the network, we argue that different layers and datasets require different receptive fields. We use multi-scale adaptive convolution kernels and dilation rates to optimize traditional multi-scale temporal convolution with a simple and effective self attention mechanism, allowing different network layers to adaptively select convolution kernels of different sizes and dilation rates instead of being fixed and unchanged. Besides, the effective receptive field of the simple residual connection is not large, and there is a great deal of redundancy in the deep residual network, which will lead to the loss of context when aggregating spatio-temporal information. This article introduces a feature fusion mechanism that replaces the residual connection between initial features and temporal module outputs, effectively solving the problems of context aggregation and initial feature fusion. We propose a multi-modality adaptive feature fusion framework (MMAFF) to simultaneously increase the receptive field in both spatial and temporal dimensions. Concretely, we input the features extracted by the spatial module into the adaptive temporal fusion module to simultaneously extract multi-scale skeleton features in both spatial and temporal parts. In addition, based on the current multi-stream approach, we use the limb stream to uniformly process correlated data from multiple modalities. Extensive experiments show that our model obtains competitive results with state-of-the-art methods on the NTU-RGB+D 60 and NTU-RGB+D 120 datasets. [ABSTRACT FROM AUTHOR]

Text image recognition in natural scenes is challenging in computer vision, even though it is already widely used in real-life applications. With the development of deep learning, the accuracy of scene text recognition has been continuously improved. The encoder-decoder architecture is currently a general framework in scene text recognition. With 2D attention on the decoder, better attention can be paid to the position of each character. However, many methods based on the encoder-decoder architecture only adopt the attention mechanism on the decoder. Therefore, their ability to locate characters is limited. In order to solve this problem, we propose a Transformer-based encoder-decoder structure with a two-stage attention mechanism for scene text recognition. At the encoder, a first-stage attention module integrating spatial attention and channel attention is used to capture the overall location of the text in the image, while at the decoder, a second-stage attention module is used to pinpoint the position of each character in the text image. This two-stage attention mechanism can locate the position of the text more effectively and improve recognition accuracy. Also, we design a multi-branch feature fusion module for the encoder that can fuse features from different receptive fields to obtain more robust features. We train the model on synthetic text datasets and test it on real scene text datasets. The experimental results show that our model is very competitive. [ABSTRACT FROM AUTHOR]

Text classification plays an important role in information science. In order to address the issues of low classification efficiency, low accuracy, and incomplete text feature extraction in existing classification methods, this work offers a two-channel hierarchical attention mechanism short text classification model (TCHAM). First, a layered word vector attention mechanism is developed to improve the capture of keywords and phrases. Second, the TextBERT model is applied to train the word vector representation to solve the problem of multiple meanings of a word. Third, a two-channel neural network is utilized to achieve parallel acceleration. Finally, the output information of the two-channel neural network is fused to raise the accuracy of news text classification. The experimental results show that under the same environment and dataset, TCHAM increases the accuracy of text classification, reaching 98.03 % for the THUCNews dataset and 95.65 % for the SogouNews dataset, and its classification performance outperforms the comparison model. [ABSTRACT FROM AUTHOR]

Solving the problem that features prediction of self-supervised monocular depth estimation is still ambiguous in low-texture regions and boundaries. We proposed an innovative self-supervised monocular depth estimation method, monocular depth self-supervised network, which integrates three effective strategies to construct an innovative self-supervised monocular depth estimation framework: (1) the attention mechanism and feature fusion module are adopted to enhance the semantic and spatial information of feature images, (2) the threshold segmentation mask is utilized to solve object motion and low-texture regions to increase image details, and (3) the residual pose module and deep reconstruction loss are used to enhance the feature extraction capability of the model to improve the accuracy of depth and pose estimation. Comprehensive experiments and visual analysis results demonstrate the effectiveness of each component in isolation. Compared to existing self-supervised methods, our model not only achieves outstanding results on KITTI and NYU Depth V2 datasets but also can be suitable to different environments. [ABSTRACT FROM AUTHOR]

Synthetic Aperture Radar (SAR), an active remote sensing imaging radar technology, has certain surface penetration ability and can work all day and in all weather conditions. It is widely applied in ship detection to quickly collect ship information on the ocean surface from SAR images. However, the ship SAR images are often blurred, have large noise interference, and contain more small targets, which pose challenges to popular one-stage detectors, such as the single-shot multi-box detector (SSD). We designed a novel network structure, a combinational fusion SSD (CF-SSD), based on the framework of the original SSD, to solve these problems. It mainly includes three blocks, namely a combinational fusion (CF) block, a global attention module (GAM), and a mixed loss function block, to significantly improve the detection accuracy of SAR images and remote sensing images and maintain a fast inference speed. The CF block equips every feature map with the ability to detect objects of all sizes at different levels and forms a consistent and powerful detection structure to learn more useful information for SAR features. The GAM block produces attention weights and considers the channel attention information of various scale feature information or cross-layer maps so that it can obtain better feature representations from the global perspective. The mixed loss function block can better learn the positions of the truth anchor boxes by considering corner and center coordinates simultaneously. CF-SSD can effectively extract and fuse the features, avoid the loss of small or blurred object information, and precisely locate the object position from SAR images. We conducted experiments on the SAR ship dataset SSDD, and achieved a 90.3% mAP and fast inference speed close to that of the original SSD. We also tested our model on the remote sensing dataset NWPU VHR-10 and the common dataset VOC2007. The experimental results indicate that our proposed model simultaneously achieves excellent detection performance and high efficiency. [ABSTRACT FROM AUTHOR]