Results 31 to 40 of about 1,512,599 (152)

Semantic-Aware Local-Global Vision Transformer [PDF]

, 2022
Vision Transformers have achieved remarkable progresses, among which Swin Transformer has demonstrated the tremendous potential of Transformer for vision tasks.
Zhang, Jiatong, Lu, Guangming, Pei, Wenjie, Chen, Fanglin, Yao, Zengwei   +4 more
core  

QuadTree Attention for Vision Transformers

CoRR, 2022
Transformers have been successful in many vision tasks, thanks to their capability of capturing long-range dependency. However, their quadratic computational complexity poses a major obstacle for applying them to vision tasks requiring dense predictions, such as object detection, feature matching, stereo, etc.
Tang, Shitao, Zhang, Jiahui, Zhu, Siyu, Tan, Ping   +3 more
openaire   +4 more sources

Lightening-Transformer

, 2023
[HPCA24] Lightening-Transformer: A Dynamically-operated Optically-interconnected Photonic Transformer ...
Hanqing Zhu (17448903)
core   +1 more source

Towards Robust Vision Transformer [PDF]

, 2022
Recent advances on Vision Transformer (ViT) and its improved variants have shown that self-attention-based networks surpass traditional Convolutional Neural Networks (CNNs) in most vision tasks.
Mao, Xiaofeng, Xue, Hui, He, Yuan, Chen, Yuefeng, Li, Xiaodan, Ye, Shaokai, Duan, Ranjie, Qi, Gege   +7 more
core  

Pyramid Vision Transformer: A Versatile Backbone for Dense Prediction without Convolutions

, 2021
Although convolutional neural networks (CNNs) have achieved great success in computer vision, this work investigates a simpler, convolution-free backbone network use-fid for many dense prediction tasks.
Deng-Ping Fan, Wang, W, Luo, P, Enze Xie, Ping Luo, Tong Lu, Ding Liang, Ling Shao, Xiang Li, Song, K, Fan, DP, Shao, L, Lu, T, Kaitao Song, Li, X, Liang, D, Wenhai Wang, Xie, E   +17 more
core   +1 more source

Optimal Topology of Vision Transformer for Real-Time Video Action Recognition in an End-To-End Cloud Solution

Machine Learning and Knowledge Extraction, 2023
This study introduces an optimal topology of vision transformers for real-time video action recognition in a cloud-based solution. Although model performance is a key criterion for real-time video analysis use cases, inference latency plays a more ...
Saman Sarraf, Milton Kabia
doaj   +1 more source

AxWin Transformer: A Context-Aware Vision Transformer Backbone with Axial Windows [PDF]

, 2023
Recently Transformer has shown good performance in several vision tasks due to its powerful modeling capabilities. To reduce the quadratic complexity caused by the attention, some outstanding work restricts attention to local regions or extends axial ...
Lin, Fangjian, Yu, Long, Ma, Yizhe, Wu, Sitong, Tian, Shengwei   +4 more
core  

$E(2)$-Equivariant Vision Transformer [PDF]

, 2023
Vision Transformer (ViT) has achieved remarkable performance in computer vision. However, positional encoding in ViT makes it substantially difficult to learn the intrinsic equivariance in data.
Yang, Kaifan, He, Fengxiang, Xu, Renjun, Liu, Ke   +3 more
core   +1 more source

V-LTCS: Backbone exploration for Multimodal Misogynous Meme detection

Natural Language Processing Journal
Memes have become a fundamental part of online communication and humour, reflecting and shaping the culture of today’s digital age. The amplified Meme culture is inadvertently endorsing and propagating casual Misogyny. This study proposes V-LTCS (Vision-
Sneha Chinivar, Roopa M.S., Arunalatha J.S., Venugopal K.R.   +3 more
doaj   +1 more source

Privacy-Preserving Semantic Segmentation Using Vision Transformer

Journal of Imaging, 2022
In this paper, we propose a privacy-preserving semantic segmentation method that uses encrypted images and models with the vision transformer (ViT), called the segmentation transformer (SETR).
Hitoshi Kiya, Teru Nagamori, Shoko Imaizumi, Sayaka Shiota   +3 more
doaj   +1 more source