Results 91 to 100 of about 18,617 (276)

Diversified Composite Prompting to Enhance Generalisation of Vision‐Language Models

CAAI Transactions on Intelligence Technology, EarlyView.
ABSTRACT In recent years, prompt learning has shown promise in transferring pretrained vision‐language models (VLMs) to downstream tasks. However, existing methods face two challenges in improving generalisation: (1) When leveraging the collaborative effect of multimodal prompts, it is often assumed that text and visual modalities share the same prompt
Xiaoyong Mei, Chong Tang, Zhengqun Dai, Fudan Zheng, Kongwen Zhang, Tianyu Lin   +5 more
wiley   +1 more source

Precipitation Nowcasting With Spatial And Temporal Transfer Learning Using Swin-UNETR

, 2023
Climate change has led to an increase in frequency of extreme weather events. Early warning systems can prevent disasters and loss of life. Managing such events remain a challenge for both public and private institutions.
Kumar, Ajitabh
core  

Classifying Deepfakes Using Swin Transformers

CoRR
3 ...
Aprille J. Xi, Eason Chen
openaire   +2 more sources

Multiple Transformer‐Based Models for Phase‐Sensitive Masking‐Based Single‐Channel Speech Enhancement

CAAI Transactions on Intelligence Technology, EarlyView.
ABSTRACT Leveraging both global contextual dependencies and local temporal‐spectral patterns can further enhance speech quality and intelligibility, motivating the integration of diverse neural network structures for improved mask estimation. Furthermore, due to the limitations of existing time‐frequency phase‐aware masks, a new constrained phase ...
Matin Pashaian, Sanaz Seyedin
wiley   +1 more source

Frequency Domain and Gradient-Spatial Multi-Scale Swin KANsformer for Remote Sensing Scene Classification

Remote Sensing
Transformer-based deep learning techniques have recently shown outstanding potential in remote sensing scene classification (RSSC), benefiting from their ability to capture global semantic relationships and contextual dependencies.
Xiaozhang Zhu, Junqing Huang, Haihui Wang   +2 more
doaj   +1 more source

Unsupervised Low Light Image Enhancement Using SNR-Aware Swin Transformer

, 2023
Image captured under low-light conditions presents unpleasing artifacts, which debilitate the performance of feature extraction for many upstream visual tasks.
Gao, Yanzeng, Hou, Yueen, Huang, Zihan, Luo, Zhijian, Tang, Jiahui   +4 more
core  

Multi‐angle, cross‐domain fusion strategy enhances automated insect identification and hierarchical categorization: a case study on assassin bugs (Hemiptera: Reduviidae)

Cladistics, EarlyView.
Abstract Automated insect identification systems hold significant value for biodiversity monitoring, pest management, citizen science initiatives and systematic studies, particularly in an era of declining expertise in insect taxonomy. However, current deep learning approaches often rely on standardized specimen photos from limited‐angles and ...
Xinkai Wang, Huaiyu Liu, Zhuo Chen, Yisheng Zhao, Yingqi Liu, Haoyang Xiong, Yuange Duan, Fang Song, Wanzhi Cai, Xuankun Li, Hu Li   +10 more
wiley   +1 more source

A Unified Deep Learning Framework for Instance Segmentation Across Diverse Cytological Stains

Cytopathology, EarlyView.
Transformer‐based unified cytology segmentation across Papanicolaou, Feulgen and AgNOR achieves stain‐invariant performance. Mask2Former maximises boundary precision (AP75) on the combined dataset, enabling one model to replace multiple stain‐specific deployments without accuracy loss while simplifying clinical integration.
Luís Otávio Santos, Rodrigo de Paula E. Silva Ribeiro, Bibiana Quatrin Tiellet, Laura Maria Effting Silva, Fabiana Botelho de Miranda Onofre, Alexandre Sherlley Casimiro Onofre, Aldo von Wangenheim   +6 more
wiley   +1 more source

FML-Swin: An Improved Swin Transformer Segmentor for Remote Sensing Images

IEEE Access
Semantic segmentation of urban remote sensing images is a very challenging task. Due to the complex background, occlusion overlap and small scale target of urban remote sensing image, the semantic segmentation results have some defects such as target confusion and similarity, target boundary ambiguity, and small scale target omission.
Tianren Wu, Wenqin Deng, Rui Lin, Junzhe Jiang 0004, Xueyun Chen   +4 more
openaire   +2 more sources

Application of Noise2Inverse and adaptation (Noise2Phase) to single‐mask x‐ray phase contrast micro‐computed tomography

Journal of Microscopy, EarlyView.
Abstract X‐ray phase contrast imaging (XPCI), when implemented in micro‐computed tomography (micro‐CT) mode, offers high‐contrast 3D imaging of weakly‐attenuating material samples. In the so‐called single‐mask edge illumination approach, a mask with periodically spaced transmitting apertures is used to split the x‐ray beam into narrow beamlets; when ...
Khushal Shah, Grammatiki Lioliou, Dai Chen, Alexander Denker, Peter Munro, Marco Endrizzi, Alberto Astolfo, Alessandro Olivo, Charlotte K. Hagen   +8 more
wiley   +1 more source