YOLOv5 Improvement Series (19) - Swin TransformerV1 to replace the backbone network (ViT model with smaller parameters) - Code World

YOLOv5 Improvement Series (19) - Swin TransformerV1 to replace the backbone network (ViT model with smaller parameters)

Enterprise 2023-08-08 20:20:32 views: null

NoSuchKey

Guess you like

Origin blog.csdn.net/weixin_43334693/article/details/132161488

YOLOv5 Improvement Series (19) - Swin TransformerV1 to replace the backbone network (ViT model with smaller parameters)

YOLOv5 Improvement Series (21) - RepViT to replace the backbone network (Tsinghua ICCV 2023|The latest open source mobile terminal ViT)

YOLOv5 Improvement Series (23) - MobileViTv2 to replace the backbone network (an efficient separable self-attention mechanism for mobile vision Transformer)

YOLOv5 improvement series (24) - MobileViTv3 that replaces the backbone network (a further upgrade of the mobile lightweight network)

Practical improvement of YOLOv8 | Replacement of the backbone network Backbone (6) Lightweight model VanillaNet advanced chapter

YOLOV5 improvement: CVPR2023: join the EfficientViT backbone: memory access efficient ViT with cascade group attention

Practical improvement of YOLOv8 | Replacing the backbone network Backbone with the 2023 latest model LSKNet, rotating target detection SOTA

YOLOv8/YOLOv7/YOLOv5/YOLOv4/Faster-rcnn series algorithm improvements [NO.83] Change the backbone feature extraction network Backbone to RevCol

YOLOv5 improvement series (13) - replace the activation function of SiLU, ReLU, ELU, Hardswish, Mish, Softplus, AconC series, etc.

YOLOv8's latest improvement series: YOLOv8+RepViT, re-examines mobile CNN from the perspective of ViT, effectively improving the model detection effect! ! ! 1.3ms delay-Tsinghua ICCV 2023 latest open source mobile network architecture RepViT

YOLOv5 improvement series (18) - replace Neck's AFPN (new progressive feature pyramid | beyond PAFPN | measured increase point)

Practical improvement of YOLOv8 | Replacing the backbone network Backbone (5) Lightweight 2023 latest network VanillaNet, the power of minimalism in deep learning

[Lightweight] Practical improvement of YOLOv8 | Replacing the backbone network Backbone's Ghostnet

[Lightweight] Practical improvement of YOLOv8 | Replacing the backbone network Backbone with GhostnetV2

Yolov5 Series 1---Yolo development history and detailed explanation of Yolov5 model

[YOLOv7/YOLOv5 series improvement NO.53] Integrating the ECVBlock module in the CFPNet network to improve the detection ability of small targets

Improving the YOLO series | YOLOv5/v7 replacing the backbone network with ResNet50/ResNet101

[YOLO series] YOLOv5, YOLOX, YOOv6, YOLOv7 network model structure

5. Kick into ViT——Swin Transformer (Part 1)

YOLOv8 latest improvement series: YOLOV8 backbone improvements - Huawei Noah proposes a new backbone architecture VanillaNet. YOLOv8 integrates the power of deep learning minimalism to greatly improve the robustness of the model! !

YOLOv8/YOLOv7/YOLOv5/YOLOv4/Faster-rcnn series algorithm improvement [NO.73] Adding asymptotic feature pyramid network (AFPN module)

Introduction to vit network model

ViT: Visual Transformer backbone network ViT paper and code detailed

ViT: Visual Transformer backbone network ViT paper and code detailed

YOLOv6 Pro | Make it easier to build a network and replace modules with YOLOv6, and help improve the network structure in scientific research, including Backbone, Neck, DecoupleHead (refer to the way YOLOv5 builds a network)

[YOLOv7/YOLOv5 series algorithm improvement NO.7] Loss function improvement

Reproduction of YoLoV5 backbone based on Pytorch (Part 1)

Swin Transformer as the JDE of the backbone network (VisDrone2019-MOT dataset)

YOLOV5/YOLOV8 improvement: CVPR23: join the ConvNeXt V2 backbone, effectively increase points

[YOLOv8/YOLOv7/YOLOv5/YOLOv4/Faster-rcnn series algorithm improvement NO.65] CVPR 2023 | Tsinghua team's plug-and-play network architecture—Slide-Transformer

Recommended

Ranking

css + html achieve 3D photo wall

Python Concise Guide: Novice will learn object-oriented []

ES6 inheritance (review prototype chain inheritance)

"A long article teaches you how to use appium in all aspects"

The third individual work - prototyping

HTML entity characters

Django (three) RESTFul of Django

Analysis of U disk file system (take FAT32 as an example)

Commonly used image drawing online experimental level - Level 5: Pie chart drawing

java programming design ideas

Daily

More

2025-05-02(0)

2025-05-01(0)

2025-04-30(0)

2025-04-29(0)

2025-04-28(0)

2025-04-27(0)

2025-04-26(0)

2025-04-25(0)

2025-04-24(0)

2025-04-23(0)