This paper focuses on the visible-thermal cross-modality person re-identification (VT Re-ID) task, whose goal is to match person images between the daytime visible modality and the nighttime thermal modality. The two-stream network is usually adopted to address the cross-modality discrepancy, the most challenging problem for VT Re-ID, by learning the multi-modality person features. In this paper, we explore how many parameters a two-stream network should share, which is still not well investigated in the existing literature. By splitting the ResNet50 model to construct the modality-specific feature extraction network and modality-sharing feature embedding network, we experimentally demonstrate the effect of parameter sharing of two-stream network for VT Re-ID. Moreover, in the framework of part-level person feature learning, we propose the hetero-center triplet loss to relax the strict constraint of traditional triplet loss by replacing the comparison of the anchor to all the other samples by the anchor center to all the other centers. With extremely simple means, the proposed method can significantly improve the VT Re-ID performance. The experimental results on two datasets show that our proposed method distinctly outperforms the state-of-the-art methods by large margins, especially on the RegDB dataset achieving superior performance, rank1/mAP/mINP 91.05%/83.28%/68.84%. It can be a new baseline for VT Re-ID, with a simple but effective strategy.

Salient object detection (SOD) aims to identify the most prominent regions in images. However, the large model sizes , high computational costs , and slow inference speeds of existing RGB-D SOD models have hindered their deployment on real-world embedded devices. To address this issue, we propose a novel method named AirSOD, which is committed to lightweight RGB-D SOD. Specifically, we first design a hybrid feature extraction network, which includes the first three stages of MobileNetV2 and our Parallel Attention-Shift convolution (PAS) module. Using the novel PAS module enables capturing both long-range dependencies and local information to enhance the representation learning while significantly reducing the number of parameters and computational complexity. Secondly, we propose a Multi-level and Multi-modal feature Fusion (MMF) module to facilitate feature fusion, and a Multi-path enhancement for Feature Refinement (MFR) decoder for feature integration. The proposed method significantly reduces the model size by 63%, decreases the computational complexity by 43%, and improves the inference speed by 43% compared with the cutting-edge model (MobileSal). We test our AirSOD on six widely-used RGB-D SOD datasets. Extensive experimental results demonstrate that our method obtains satisfactory performance. The source codes will be made available.

Laryngeal cancer tumor (LCT) grading is a challenging task in P63 Immunohistochemical (IHC) histopathology images due to small differences between LCT levels in pathology images, the lack of precision in lesion regions of interest (LROIs) and the paucity of LCT pathology image samples. The key to solving the LCT grading problem is to transfer knowledge from other images and to identify more accurate LROIs, but the following problems occur: 1) transferring knowledge without a priori experience often causes negative transfer and creates a heavy workload due to the abundance of image types, and 2) convolutional neural networks (CNNs) constructing deep models by stacking cannot sufficiently identify LROIs, often deviate significantly from the LROIs focused on by experienced pathologists, and are prone to providing misleading second opinions. So we propose a novel fusion attention block network (FABNet) to address these problems. First, we propose a model transfer method based on clinical a priori experience and sample analysis (CPESA) that analyzes the transfer ability by integrating clinical a priori experience using indicators such as the relationship between the cancer onset location and morphology and the texture and staining degree of cell nuclei in histopathology images; our method further validates these indicators by the probability distribution of cancer image samples. Then, we propose a fusion attention block (FAB) structure, which can both provide an advanced non-uniform sparse representation of images and extract spatial relationship information between nuclei; consequently, the LROI can be more accurate and more relevant to pathologists. We conducted extensive experiments, compared with the best Baseline model, the classification accuracy is improved 25%, and It is demonstrated that FABNet performs better on different cancer pathology image datasets and outperforms other state of the art (SOTA) models.

Cervical cancer is the fourth most common cancer in the world. Whole-slide images (WSIs) are an important standard for the diagnosis of cervical cancer. Missed diagnoses and misdiagnoses often occur due to the high similarity in pathological cervical images, the large number of readings, the long reading time, and the insufficient experience levels of pathologists. Existing models have insufficient feature extraction and representation capabilities, and they suffer from insufficient pathological classification. Therefore, this work first designs an image processing algorithm for data augmentation. Second, the deep convolutional features are extracted by fine-tuning pre-trained deep network models, including ResNet50 v2, DenseNet121, Inception v3, VGGNet19, and Inception-ResNet, and then local binary patterns and a histogram of the oriented gradient to extract traditional image features are used. Third, the features extracted by the fine-tuned models are serially fused according to the feature representation ability parameters and the accuracy of multiple experiments proposed in this paper, and spectral embedding is used for dimension reduction. Finally, the fused features are inputted into the Analysis of Variance-F value-Spectral Embedding Net (AF-SENet) for classification. There are four different pathological images of the dataset: normal, low-grade squamous intraepithelial lesion (LSIL), high-grade squamous intraepithelial lesion (HSIL), and cancer. The dataset is divided into a training set (90%) and a test set (10%). The serial fusion effect of the deep features extracted by Resnet50v2 and DenseNet121 (C₅) is the best, with average classification accuracy reaching 95.33%, which is 1.07% higher than ResNet50 v2 and 1.05% higher than DenseNet121. The recognition ability is significantly improved, especially in LSIL, reaching 90.89%, which is 2.88% higher than ResNet50 v2 and 2.1% higher than DenseNet121. Thus, this method significantly improves the accuracy and generalization ability of pathological cervical WSI recognition by fusing deep features.

This paper focuses on the high-resolution (HR) remote sensing images semantic segmentation task, whose goal is to predict semantic labels in a pixel-wise manner. Due to the rich complexity and heterogeneity of information in HR remote sensing images, the ability to extract spatial details (boundary information) and semantic context information dominates the performance in segmentation. In this paper, based on the frequently used fully convolutional network framework, we propose a boundary enhancing semantic context network (BES-Net) to explicitly use the boundary to enhance semantic context extraction. BES-Net mainly consists of three modules: (1) a boundary extraction module for extracting the semantic boundary information, (2) a multi-scale semantic context fusion module for fusing semantic features containing objects with multiple scales, and (3) a boundary enhancing semantic context module for explicitly enhancing the fused semantic features with the extracted boundary information to improve the intra-class semantic consistency, especially in those pixels containing boundaries. Extensive experimental evaluations and comprehensive ablation studies on the ISPRS Vaihingen and Potsdam datasets demonstrate the effectiveness of BES-Net, yielding an overall improvement of 1.28/2.36/0.72 percent in mF1/mIoU/OA over FCN_8s when the BE and MSF modules are combined by the BES module. In particular, our BES-Net achieves a state-of-the-art performance of 91.4% OA on the ISPRS Vaihingen dataset and 92.9%/91.5% mF1/OA on the ISPRS Potsdam dataset.

The performance of a deep learning-based Synthetic aperture radar (SAR) automatic target recognition (ATR) model largely relies on the scale and quality of training samples. However, it is time-consuming and expensive to collect sufficient data in practice. Although generative adversarial network (GAN) provides a way for SAR target image generation, existing GAN-based methods cannot confirm what features the generator learns, thus they struggle in generating precise SAR target images. In this paper, we propose an angle transformation GAN (ATGAN) that can generate azimuth-controllable SAR target images while preserving the target details. The key idea of our ATGAN is to reframe the generation task from the perspective of image-to-image translation. To this end, ATGAN consists of two modules, a coarse-to-fine generator that aims to learn the angle transformation in the deep feature space and then apply it to manipulate the representation of an input SAR target image to generate a new one, while a spectral-normalized patch discriminator that tries to estimate the probability that an input SAR target image is real rather than fake using a patch-averaged strategy. Combining with spatial transformer and adversarial training paradigm, ATGAN can generate precise SAR target images for ATR. Extensive experiments verify the effectiveness of the proposed ATGAN, and our method outperforms the state-of-the-art method qualitatively and quantitatively.