相关参考文献
注意:仅列出部分参考文献,下载原文获取全部文献信息。
Article
Computer Science, Artificial Intelligence
Wei Tang et al.
Summary: This paper proposes an infrared and visible image fusion method based on Transformer and cross-correlation, called TCCFusion. The method uses a local feature extraction branch and a global feature extraction branch to preserve local and global useful information. A cross-correlation loss is used to train the fusion model. Experimental results show that TCCFusion outperforms state-of-the-art algorithms in both visual quality and quantitative assessments.
PATTERN RECOGNITION
(2023)
Article
Robotics
Mingjian Liang et al.
Summary: RGB-Thermal based perception has made significant progress. Thermal information is valuable when visual cameras face poor lighting conditions. However, effectively fusing RGB images and thermal data remains a challenge.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2023)
Article
Geosciences, Multidisciplinary
Qian Shi et al.
Summary: A general deep learning framework is proposed for large-scale urban green space (UGS) mapping, and fine-grained UGS maps of 31 major cities in mainland China are generated. The framework includes a generator and a discriminator, and incorporates pre-training and adversarial training for model performance improvement and adaptation to different cities. Evaluation results show that the framework achieves high accuracy and F1 score.
EARTH SYSTEM SCIENCE DATA
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Wenda Zhao et al.
Summary: This paper proposes a meta-feature embedding method for infrared and visible image fusion, which is optimized by simulating meta learning. It also implements mutual promotion learning between fusion and detection tasks. Experimental results demonstrate the effectiveness of the proposed method.
2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION (CVPR)
(2023)
Proceedings Paper
Computer Science, Artificial Intelligence
Jiaming Zhang et al.
Summary: Multimodal fusion can enhance the robustness of semantic segmentation. However, there is limited research on fusing an arbitrary number of modalities. To address this issue, the DELIVER benchmark dataset is created, which includes Depth, LiDAR, multiple Views, Events, and RGB modalities. The proposed CMNEXT model achieves state-of-the-art performance on various datasets and allows for scaling from 1 to 81 modalities.
2023 IEEE/CVF CONFERENCE ON COMPUTER VISION AND PATTERN RECOGNITION, CVPR
(2023)
Article
Environmental Sciences
Qingwang Wang et al.
Summary: In the field of remote sensing image applications, object detection using RGB and infrared images is an important technology. This study proposes a redundant information suppression network (RISNet) to enhance the fusion of complementary information between RGB and infrared images. Experimental results show that the proposed method outperforms state-of-the-art approaches, especially in challenging conditions.
Article
Geography, Physical
Da He et al.
Summary: This study addresses the limitations of existing land cover products by proposing a progressive edge-guided super-resolution architecture and an alternating optimization strategy, providing a new approach and perspective for observing urban dynamics and their underlying mechanisms.
GISCIENCE & REMOTE SENSING
(2022)
Article
Computer Science, Artificial Intelligence
Zongbo Han et al.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2022)
Article
Computer Science, Artificial Intelligence
Yang Zhang et al.
IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE
(2020)
Article
Robotics
Yuxiang Sun et al.
IEEE ROBOTICS AND AUTOMATION LETTERS
(2019)