Results 71 to 80 of about 1,512,599 (152)

FSwin Transformer: Feature-Space Window Attention Vision Transformer for Image Classification

IEEE Access
The vision transformer (ViT) with global self-attention exhibits quadratic computational complexity that depends on the image size. To address this issue, window-based self-attention ViT limits attention area to a specific window, thereby mitigating the ...
Dayeon Yoo, Jeesu Kim, Jinwoo Yoo
doaj   +1 more source

A customised vision transformer for accurate detection and classification of Java Plum leaf disease [PDF]

Vision Transformer (ViT) has recently attracted significant attention for its performance in image classification. However, studies have yet to explore its potential in detecting and classifying plant leaf disease.
Bhowmik, Auvick Chandra, Emon, Yousuf Rayhan, Ahad, Md. Taimur, Ahmed, Faruk, Li, Yan, Song, Bo   +5 more
core   +1 more source

Unsupervised Pose Estimation by Means of an Innovative Vision Transformer

, 2023
Attention-only Transformers [34] have been applied to solve Natural Language Processing (NLP) tasks and Computer Vision (CV) tasks. One particular Transformer architecture developed for CV is the Vision Transformer (ViT) [15].
Russo S., Brandizzi N., Fanti A., Iocchi L., Gallotta R., Nardi D., Napoli C., Andrea Fanti, Luca Iocchi, Daniele Nardi, Nicolo’ Brandizzi, Roberto Gallotta, Christian Napoli, Samuele Russo   +13 more
core   +1 more source

Vision Transformer With Quadrangle Attention

IEEE Transactions on Pattern Analysis and Machine Intelligence
Window-based attention has become a popular choice in vision transformers due to its superior performance, lower computational complexity, and less memory footprint. However, the design of hand-crafted windows, which is data-agnostic, constrains the flexibility of transformers to adapt to objects of varying sizes, shapes, and orientations.
Qiming Zhang 0001, Jing Zhang 0037, Yufei Xu, Dacheng Tao   +3 more
openaire   +4 more sources

MoR–Swin: Efficient Vision Transformer Using Mixture of Recursions

Journal of Imaging
Vision Transformers, especially Swin Transformer, have become default backbones for various vision tasks but suffer from high memory consumption and training costs. This letter proposes MoR–Swin, a novel architecture that integrates Mixture of Recursions
Yongbao Ai, Tianxiang Gao, Zhipeng Lin, Longqi Yang, Qingyu Chang   +4 more
doaj   +1 more source

Attacking Compressed Vision Transformers

, 2023
Vision Transformers are increasingly embedded in industrial systems due to their superior performance, but their memory and power requirements make deploying them to edge devices a challenging task. Hence, model compression techniques are now widely used to deploy models on edge devices as they decrease the resource requirements and make model ...
Swapnil Parekh, Devansh Shah, Pratyush Shukla   +2 more
openaire   +3 more sources

Biomedical image classification using vision transformer [PDF]

E3S Web of Conferences
Particularly in the areas of public health and welfare, biomedical image classification plays a significant role in advancing the Sustainable Development Goals (SDGs).
Tjahyaningtijas Hapsari Peni Agustin, Nashrullah Muhammad Aamir, Puspitaningayu Pradini, Rakhmawati Lusia, Yamasari Yuni, Paragas Jessie R.   +5 more
doaj   +1 more source

Classification of Mobile-Based Oral Cancer Images Using the Vision Transformer and the Swin Transformer

Oral cancer, a pervasive and rapidly growing malignant disease, poses a significant global health concern. Early and accurate diagnosis is pivotal for improving patient outcomes.
Bofan Song, Dharma Raj KC, Shaobai Li, Rubin Yuchan Yang, Rongguang Liang, Chicheng Zhang   +5 more
core   +1 more source

Vision transformer for contrastive clustering

Knowledge-Based Systems
Vision Transformer (ViT) has shown its advantages over the convolutional neural network (CNN) with its ability to capture global long-range dependencies for visual representation learning. Besides ViT, contrastive learning is another popular research topic recently. While previous contrastive learning works are mostly based on CNNs, some recent studies
Hua-Bao Ling, Bowen Zhu, Dong Huang 0001, Ding-Hua Chen, Chang-Dong Wang 0001, Jian-Huang Lai   +5 more
openaire   +3 more sources

Frequency-aware vision transformers for high-fidelity super-resolution of Earth system models

Scientific Reports
Super-resolution can play an essential role in enhancing the spatial fidelity of Earth System Model outputs, allowing fine-scale structures highly beneficial to climate science to be recovered from coarse simulations.
Ehsan Zeraatkar, Salah A. Faroughi, Jelena Tešić   +2 more
doaj   +1 more source