Single-frame infrared small target detection is still a challenging task due to the complex background and unobvious structural characteristics of small targets. Recently, convolutional neural networks (CNN) began to appear in the field of infrared small target detection and have been widely used for excellent performance. However, existing CNN-based methods mainly focus on local spatial features while ignoring the long-range contextual dependencies between small targets and backgrounds. To capture the global context-aware information, we propose fusion network architecture of transformer and CNN (FTC-Net), which consists of two branches. The CNN-based branch uses a U-Net with skip connections to obtain low-level local details of small targets. The transformer-based branch applies hierarchical self-attention mechanisms to learn long-range contextual dependencies. Specifically, the transformer branch can suppress background interferences and enhance target features. To obtain local and global feature representation, we design a feature fusion module to realize the feature concentration of two branches. We implement ablation and comparative experiments on a publicly accessed SIRST dataset. Experimental results show that the transformer-based branch is effective and suggest the superiority of the proposed FTC-Net compared with other state-of-the-art methods.

Person attribute recognition aims to identify the attribute labels from the pedestrian images. Extracting contextual relation from the images and attributes, including the spatial-semantic relations, the spatial context and the semantic correlation, is beneficial to enhance the discrimination of the features for recognizing the attributes. Thus, this work proposes a sequence contextual relation learning (SCRL) method to capture these relations. It first embeds the images and attributes into sequences in two branches. Then SCRL flexibly learns the contextual relation from the sequences with the parallel attention model structure, which integrates the inter-attention and intra-attention models. The inter-attention module is utilized to extract the spatial-semantic relations, while the intra-attention is designed to gain the spatial context and the semantic correlation. Both attention modules are comprised of several parallel attention units and each unit can obtain the pairwise relations in one subspace. Therefore, they obtain the relations in multiple subspaces, which can improve the comprehensiveness of the relation learning. Additionally, for the sake of better extraction of spatial-semantic relations, this paper employs connectionist temporal classification (CTC) loss which is capable of driving the network to enforce monotonic alignment between the image and attribute. It can also accelerate the convergence of the network by the algorithm in it. Extensive experiments on five public datasets, i.e., Market-1501 attribute, Duke attribute, PETA, RAP and PA-100K datasets, demonstrate the effectiveness of the proposed method.

Infrared-Visible person Re-IDentification(IV-ReID) is an emerging subject, which has important research significance for nighttime monitoring. Existing works focus on reducing cross-modality discrepancies, but the cross-modality discrepancy cannot be completely eliminated. Therefore, we concentrate on excavating cross-modality commonalities to handle the task. Since similar features between two modalities are possessed of cross-modality commonalities, our goal is to find more similar features in infrared and visible images. A novel Dual-stream Multi-layer Corresponding Fusion Network(DMCF) is proposed to explore more similar features between two modalities in this paper. It mainly contains three aspects. 1) We explore more similar features between two modalities by learning low-level features, Meanwhile, we also propose a method that the same level features between two modalities are correspondingly fused to reduce cross-modality discrepancies. 2) We adopt different Multi-granularity dividing methods for multi-layer features, so that it can improve the ability of the model to perceive feature details. 3) We separately calculate the loss for different-layer features. Therefore, we learn different weighting factors for the loss of different hierarchical features through Multi-task Learning, so that each branch can be fully optimized. Extensive experiments on two datasets demonstrate the superior performance compared to the state-of-the-arts.

The Infrared Object Tracking (IOT) task aims to locate objects in infrared sequences. Since color and texture information is unavailable in infrared modality, most existing infrared trackers merely rely on capturing spatial contexts from the image to enhance feature representation, where other complementary information is rarely deployed. To fill this gap, we in this article propose a novel Asymmetric Deformable Spatio-Temporal Framework (ADSF) to fully exploit collaborative shape and temporal clues in terms of the objects. Firstly, an asymmetric deformable cross-attention module is designed to extract shape information, which attends to the deformable correlations between distinct frames in an asymmetric manner. Secondly, a spatio-temporal tracking framework is coined to learn the temporal variance trend of the object during the training process and store the template information closest to the tracking frame when testing. Comprehensive experiments demonstrate that ADSF outperforms state-of-the-art methods on three public datasets. Extensive ablation experiments further confirm the effectiveness of each component in ADSF. Furthermore, we conduct generalization validation to demonstrate that the proposed method also achieves performance gains in RGB-based tracking scenarios.

Infrared small target detection remains challenging due to weak target responses, cluttered backgrounds, and large variations in target appearance. In this paper, we propose FLINet, a fuzzy-logic-guided network that introduces a fuzzy set perspective for uncertainty-aware feature fusion in infrared small target detection. FLINet follows a U-shaped encoder-decoder architecture and incorporates two key modules: a Global-Local Fuzzy computation module (GLF) and a Difference-Aware Feature fusion module (DAF). GLF performs dual-branch fuzzy reasoning at both local-window and global-image levels to model feature uncertainty during cross-level fusion. Specifically, the local branch enhances contrast-sensitive target cues via adaptive membership-based amplification, while the global branch suppresses clutter-dominant responses through global statistical normalization, jointly improving target-background separation and morphology preservation. DAF further strengthens cross-level integration by exploiting inter-level discrepancy maps to guide a self-attention-based fusion, prioritizing complementary fine-grained details from shallow layers and semantic cues from deep layers to reduce small-target information loss. Extensive experiments on three benchmark datasets (NUDT-SIRST, NUAA-SIRST, and IRSTD-1K) demonstrate that FLINet consistently achieves superior performance across multiple evaluation metrics. Our code is available at: https://github.com/PZYang/FLINet.