旋转目标检测:mmrotate仓库中 “主要模型” 及其 “配置文件” 的列表

mmrotate

目录:

  • mmrotate 仓库中的主要模型和配置
    • Background and Motivation 背景与动机
    • Methods Overview 方法概述
      • 1. CFA
        • CFA: Convex-hull Feature Adaptation for Oriented and Densely Packed Object Detection
        • CFA:用于定向和密集对象检测的凸包特征适应
      • 2. ConvNeXt
        • ConvNeXt: A ConvNet for the 2020s
        • ConvNeXt:面向2020年代的卷积神经网络
      • 3. Circular Smooth Label (CSL)
        • Circular Smooth Label: Solving Boundary Discontinuity in Oriented Object Detection
        • 圆形平滑标签:解决定向目标检测中的边界不连续性问题
      • 4. G-Reppoints
        • G-Rep: Gaussian Representation for Arbitrary-Oriented Object Detection
        • G-Rep:用于任意方向目标检测的高斯表示
      • 5. Gliding Vertex
        • Gliding Vertex: Gliding Vertex on the Horizontal Bounding Box for Multi-Oriented Object Detection
        • Gliding Vertex:在水平边界框上滑动顶点用于多方向目标检测
      • 6. GWD
        • GWD: Rethinking Rotated Object Detection with Gaussian Wasserstein Distance Loss
        • GWD:用高斯沃瑟斯坦距离损失重新思考旋转目标检测
      • 7. K-FIoU
        • KFIoU: The KFIoU Loss for Rotated Object Detection
        • KFIoU:用于旋转目标检测的KFIoU损失
      • 8. KLD
        • KLD: Learning High-Precision Bounding Box for Rotated Object Detection via Kullback-Leibler Divergence
        • KLD:通过Kullback-Leibler散度学习高精度旋转目标检测边界框
      • 9. Oriented R-CNN
        • Oriented R-CNN: Oriented R-CNN for Object Detection
        • Oriented R-CNN:用于目标检测的定向R-CNN
      • 10. Oriented Reppoints
        • Oriented Reppoints: Oriented Reppoints for Aerial Object Detection
        • Oriented Reppoints:用于航空目标检测的定向Reppoints
      • 11. R3Det
        • R3Det: Refined Single-Stage Detector with Feature Refinement for Rotating Object
        • R3Det:具有特征优化的精细单阶段旋转目标检测器
      • 12. ReDet (Rotation-equivariant Network)
        • ReDet: A Rotation-equivariant Detector for Aerial Object Detection
        • ReDet:用于航空目标检测的旋转等变检测器
      • 13. RoI Transformer
        • RoI Transformer: Learning RoI Transformer for Oriented Object Detection in Aerial Images
        • RoI Transformer:在航空图像中用于定向目标检测的RoI Transformer
      • 14. Rotated ATSS
        • Rotated ATSS: Bridging the Gap Between Anchor-based and Anchor-free Detection via Adaptive Training Sample Selection
        • Rotated ATSS:通过自适应训练样本选择弥合基于锚点和无锚点检测之间的差距
      • 15. Rotated Faster R-CNN
        • Rotated Faster R-CNN: Towards Real-Time Object Detection with Region Proposal Networks
        • Rotated Faster R-CNN:用于实时目标检测的旋转区域提议网络
      • 16. Rotated FCOS
        • Rotated FCOS: Fully Convolutional One-Stage Object Detection
        • Rotated FCOS:全卷积单阶段目标检测
      • 17. Rotated Reppoints
        • Rotated Reppoints: RepPoints for Arbitrary-Oriented Object Detection
        • Rotated Reppoints:用于任意方向目标检测的Reppoints
      • 18. Rotated RetinaNet
        • Rotated RetinaNet: Focal Loss for Dense Object Detection
        • Rotated RetinaNet:用于密集目标检测的焦点损失
      • 19. S2A-Net (Single-stage Anchor-free Network)
        • S2A-Net: Align Deep Features for Oriented Object Detection
        • S2A-Net:对齐深度特征用于定向目标检测
      • 20. SASM Reppoints
        • SASM: Shape-Adaptive Selection and Measurement for Oriented Object Detection
        • SASM:用于定向目标检测的形状自适应选择和测量
  • 官方对比

mmrotate 仓库中的主要模型和配置

Background and Motivation 背景与动机

随着遥感图像、无人驾驶和其他复杂场景中目标检测需求的增加,特别是对于任意方向和形状复杂的目标,传统的目标检测方法往往表现出局限性。mmrotate项目旨在集成多种先进的旋转目标检测方法,为研究人员和工程师提供高效、准确的解决方案。

Methods Overview 方法概述

mmrotate项目包含多种旋转目标检测模型,每种模型在处理目标的方向和形状特征方面都有独特的设计和优化策略。以下是各模型的详细描述和实验结果概述:

1. CFA

  • 目录cfa
  • 描述:待补充详细描述。

你可能感兴趣的:(AI与SLAM论文解析,旋转目标检测,深度学习,mmrotate)