RT-DETR融合YOLOv12中的R-ELAN结构

---

RT-DETR使用教程： RT-DETR使用教程 RT-DETR改进汇总贴：RT-DETR更新汇总贴

---

《YOLOv12: Attention-Centric Real-Time Object Detectors》 一、 模块介绍

        论文链接： arxiv.org/abs/2502.12524

        代码链接： gitcode /gh_mirrors/yo/yolov12

论文速览：

       长期以来，增强YOLO框架的网络架构一直至关重要，但一直专注于基于cnn的改进，尽管注意力机制在建模能力方面已被证明具有优越性。这是因为基于注意力的模型无法匹配基于cnn的模型的速度。本文提出了一种以注意力为中心的YOLO框架，即YOLOv12，与之前基于cnn的YOLO框架的速度相匹配，同时利用了注意力机制的性能优势。YOLOv12在精度和速度方面超越了所有流行的实时目标检测器。例如，YOLOv12-N在T4 GPU上以1.64ms的推理延迟实现了40.6% mAP，以相当的速度超过了高级的YOLOv10-N / YOLOv11-N 2.1%/1.2% mAP。这种优势可以扩展到其他模型规模。YOLOv12还超越了改善DETR的端到端实时检测器，如RT-DETR /RT-DETRv2: YOLOv12- s比RT-DETR- r18 / RT-DETRv2-r18运行更快42%，仅使用36%的计算和45%的参数。更多的比较见图1。

总结：本文将其中的R-ELAN思想融入其他模块。​

---

二、二创融合模块 2.1 相关二创模块及所需参数

        该模块可如图加入到RepNCSPELAN4、RepC3与自研等模块中，代码见群文件，所需参数如下。

RepNCSPELAN4-变式模块 所需参数：(c1, c2, c3, c4, n)

CCRI及变式模块 所需参数：(c1, c2, k, n, lightconv, shortcut, scale, e, act)

RepC4及变式模块 所需参数：(c1, c2, n, e)

        其中，RepNCSPELAN4模块的代码如下：

class RepNCSPELAN4_R_ELAN(nn.Module): """CSP-ELAN.""" def __init__(self, c1, c2, c3, c4, n=1): """Initializes CSP-ELAN layer with specified channel sizes, repetitions, and convolutions.""" super().__init__() self.c = c3 // 2 self.cv1 = Conv(c1, self.c, 1, 1) self.cv2 = nn.Sequential(RepCSP(c3 // 2, c4, n), Conv(c4, c4, 3, 1)) self.cv3 = nn.Sequential(RepCSP(c4, c4, n), Conv(c4, c4, 3, 1)) self.cv4 = Conv(self.c + (2 * c4), c2, 1, 1) def forward(self, x): """Forward pass through RepNCSPELAN4 layer.""" y = [self.cv1(x)] y.extend((m(y[-1])) for m in [self.cv2, self.cv3]) return self.cv4(torch.cat(y, 1)) def forward_split(self, x): """Forward pass using split() instead of chunk().""" y = list(self.cv1(x).split((self.c, self.c), 1)) y.extend(m(y[-1]) for m in [self.cv2, self.cv3]) return self.cv4(torch.cat(y, 1))

---

2.2 更改yaml文件 （以自研模型加入为例）

yam文件解读：YOLO系列 “.yaml“文件解读_yolo yaml文件-CSDN博客

       打开更改ultralytics/cfg/models/rt-detr路径下的rtdetr-l.yaml文件，替换原有模块。

​

# Ultralytics YOLO 🚀, AGPL-3.0 license # RT-DETR-l object detection model with P3-P5 outputs. For details see docs.ultralytics /models/rtdetr # ⭐⭐Powered by blog.csdn.net/StopAndGoyyy, 技术指导QQ:2668825911⭐⭐ # Parameters nc: 80 # number of classes scales: # model compound scaling constants, i.e. 'model=yolov8n-cls.yaml' will call yolov8-cls.yaml with scale 'n' # [depth, width, max_channels] l: [1.00, 1.00, 512] # n: [ 0.33, 0.25, 1024 ] # s: [ 0.33, 0.50, 1024 ] # m: [ 0.67, 0.75, 768 ] # l: [ 1.00, 1.00, 512 ] # x: [ 1.00, 1.25, 512 ] # ⭐⭐Powered by blog.csdn.net/StopAndGoyyy, 技术指导QQ:2668825911⭐⭐ backbone: # [from, repeats, module, args] - [-1, 1, Conv, [64, 3, 2]] # 0-P1/2 - [-1, 1, Conv, [128, 3, 2]] # 1-P2/4 - [-1, 2, CCRI, [128, 5, True, False]] - [-1, 1, Conv, [256, 3, 2]] # 3-P3/8 - [-1, 4, CCRI, [256, 3, True, True]] - [-1, 1, Conv, [512, 3, 2]] # 5-P4/16 - [-1, 4, RepNCSPELAN4_R_ELAN, [512, 512, 256, 1]] - [-1, 1, Conv, [1024, 3, 2]] # 7-P5/32 - [-1, 2, CCRI, [1024, 3, True, False]] head: - [-1, 1, Conv, [256, 1, 1, None, 1, 1, False]] # 9 input_proj.2 - [-1, 1, AIFI, [1024, 8]] - [-1, 1, Conv, [256, 1, 1]] # 11, Y5, lateral_convs.0 - [-1, 1, nn.Upsample, [None, 2, "nearest"]] - [6, 1, Conv, [256, 1, 1, None, 1, 1, False]] # 13 input_proj.1 - [[-2, -1], 1, Concat, [1]] - [-1, 2, RepC4, [256]] # 15, fpn_blocks.0 - [-1, 1, Conv, [256, 1, 1]] # 16, Y4, lateral_convs.1 - [-1, 1, nn.Upsample, [None, 2, "nearest"]] - [4, 1, Conv, [256, 1, 1, None, 1, 1, False]] # 18 input_proj.0 - [[-2, -1], 1, Concat, [1]] # cat backbone P4 - [-1, 2, RepC4, [256]] # X3 (20), fpn_blocks.1 - [-1, 1, Conv, [256, 3, 2]] # 22, downsample_convs.0 - [[-1, 16], 1, Concat, [1]] # cat Y4 - [-1, 2, RepC4, [256]] # F4 (23), pan_blocks.0 - [-1, 1, Conv, [256, 3, 2]] # 24, downsample_convs.1 - [[-1, 11], 1, Concat, [1]] # cat Y5 - [-1, 2, RepC4, [256]] # F5 (26), pan_blocks.1 - [[20, 23, 26], 1, RTDETRDecoder, [nc]] # Detect(P3, P4, P5) # ⭐⭐Powered by blog.csdn.net/StopAndGoyyy, 技术指导QQ:2668825911⭐⭐

---

 2.2 修改train.py文件

       创建Train_RT脚本用于训练。

from ultralytics.models import RTDETR import os os.environ['KMP_DUPLICATE_LIB_OK'] = 'True' if __name__ == '__main__': model = RTDETR(model='ultralytics/cfg/models/rt-detr/rtdetr-l.yaml') # model.load('yolov8n.pt') model.train(data='./data.yaml', epochs=2, batch=1, device='0', imgsz=640, workers=2, cache=False, amp=True, mosaic=False, project='runs/train', name='exp')

​

         在train.py脚本中填入修改好的yaml路径，运行即可训。

​

---