awesome_vision_transformer
Implementation of vision transformer. ⭐⭐⭐
https://github.com/murufeng/awesome_vision_transformer
Last synced: 17 days ago
JSON representation
Table of Contents
Awesome Survey
DETR变种
- Video Swin Transformer
- paper
- MViT: Mask Vision Transformer for Facial Expression Recognition in the wild
- CPTR
- UP-DETR
- paper
- code
- paper
- code
- paper
- code
- DA-DETR
- paper
- code
- Pointformer
- ViT-FRCNN
- Oriented Object Detection with Transformer
- paper
- code
- COTR
- paper
- code
- CAT
- M2TR
- Transformer Transforms Salient Object Detection and Camouflaged Object Detection
- SSTN
- TSP-FCOS
- ACT
- PED
- paper
- code
- Fully Transformer Networks for Semantic ImageSegmentation
- TransVOS
- paper
- code
- VisTR
- paper
- code
- https://fudan-zvg.github.io/SETR/
- https://github.com/fudan-zvg/SETR
- https://arxiv.org/abs/2012.15840
- 链接
- code
- 链接
- Code
- paper
- code
- STGT
- Transformer Tracking
- TransCenter
- TrackFormer
- paper
- code
- paper
- code
- TimeSformer
- VidTr
- ViViT
- VTN
- paper
- code
- TransPose
- TFPose
- Lifting Transformer for 3D Human Pose Estimation in Video
- paper
- code
- Vision Transformer Architecture Search
- FaceT
- TransReID
- paper
- code
- SUNETR
- U-Transformer
- paper
- code
- paper
- code
- paper
- IPT
- paper
- code
- paper
- code
- Chasing Sparsity in Vision Transformers:An End-to-End Exploration
- paper
- code
- Deepfake Video Detection Using Convolutional Vision Transformer
- Training Vision Transformers for Image Retrieval
- code
MLP
[Papers](#paper)
- RepMLP
- gMLP
- Spatial shift($S^{2}$) MLP V1
- S^2-MLPv2: Improved Spatial-Shift MLP Architecture for Vision
- GFNet | MLP领域再发力，清华大学提出将FFT思想用于空间信息交互
- CycleMLP: A MLP-like Architecture for Dense Prediction
- AS-MLP
Paper(最新，最受关注的)
- https://arxiv.org/abs/2106.12368
- CMT: Convolutional Neural Networks Meet Vision Transformers
- Early Convolutions Help Transformers See Better
- https://arxiv.org/abs/2106.13112
- https://github.com/sail-sg/volo
- https://github.com/Andrew-Qibin/VisionPermutator
- CAT: Cross Attention in Vision Transformer
- CoAtNet: Marrying Convolution and Attention for All Data Sizes
- Container: Context Aggregation Network
- Aggregating Nested Transformers
- X-volution: On the unification of convolution and self-attention
- https://github.com/Andrew-Qibin/VisionPermutator
微软Transformer霸榜模型
- paper
- DVT
- paper
- Scaling Vision Transformers
- Shuffle Transformer
- paper
- code
- paper
- code
- VTs
- paper
- code
- paper
- code
- LeViT
- CrossViT
- CeiT
- DeepViT
- paper
- paper
- code
- BoTNet
- paper
- code
- paper
- code
- paper
- code
- paper
- code
- GasHis-Transformer
- paper
- code
- RegionViT
- PVT
- code
- paper
- [paper
- code
- paper
- code
- paper
- code
- paper
- code
- paper
- code
- paper
- code
- ConViT
- paper
- code
- paper
- code
- code
- code
- code
论文解读
- 一文总结微软研究院Transformer霸榜模型三部曲
- 颜水成团队提出VOLO屠榜CV任务，无需额外训练数据，首次在ImageNet 上达到87.1%
- 分层级联Transformer！苏黎世联邦提出TransCNN: 显著降低了计算/空间复杂度！
- 登上更高峰！颜水成、程明明团队开源ViP，引入三维信息编码机制，无需卷积与注意力
- 清华鲁继文团队提出DynamicViT：一种高效的动态稀疏化Token的ViT
- 并非所有图像都值16x16个词--- 清华&华为提出一种自适应序列长度的动态ViT
- 注意力可以使MLP完全替代CNN吗？ 未来有哪些研究方向？
- 超越Swin Transformer！谷歌提出了收敛更快、鲁棒性更强、性能更强的NesT
- 超越Swin Transformer！谷歌提出了收敛更快、鲁棒性更强、性能更强的NesT
- 登上更高峰！颜水成、程明明团队开源ViP，引入三维信息编码机制，无需卷积与注意力
- 超越Swin Transformer！谷歌提出了收敛更快、鲁棒性更强、性能更强的NesT
- 登上更高峰！颜水成、程明明团队开源ViP，引入三维信息编码机制，无需卷积与注意力
- 超越Swin Transformer！谷歌提出了收敛更快、鲁棒性更强、性能更强的NesT
- 超越Swin Transformer！谷歌提出了收敛更快、鲁棒性更强、性能更强的NesT
- 一文总结微软研究院Transformer霸榜模型三部曲
- 颜水成团队提出VOLO屠榜CV任务，无需额外训练数据，首次在ImageNet 上达到87.1%
- 颜水成团队提出VOLO屠榜CV任务，无需额外训练数据，首次在ImageNet 上达到87.1%
- 颜水成团队提出VOLO屠榜CV任务，无需额外训练数据，首次在ImageNet 上达到87.1%
- 清华鲁继文团队提出DynamicViT：一种高效的动态稀疏化Token的ViT
- 并非所有图像都值16x16个词--- 清华&华为提出一种自适应序列长度的动态ViT
- 登上更高峰！颜水成、程明明团队开源ViP，引入三维信息编码机制，无需卷积与注意力
- 清华鲁继文团队提出DynamicViT：一种高效的动态稀疏化Token的ViT
- 并非所有图像都值16x16个词--- 清华&华为提出一种自适应序列长度的动态ViT
- 超越Swin Transformer！谷歌提出了收敛更快、鲁棒性更强、性能更强的NesT
- 注意力可以使MLP完全替代CNN吗？ 未来有哪些研究方向？
- 分层级联Transformer！苏黎世联邦提出TransCNN: 显著降低了计算/空间复杂度！
- 超越Swin Transformer！谷歌提出了收敛更快、鲁棒性更强、性能更强的NesT
- 注意力可以使MLP完全替代CNN吗？ 未来有哪些研究方向？
- 一文总结微软研究院Transformer霸榜模型三部曲
- 一文总结微软研究院Transformer霸榜模型三部曲