【YOLOv12多模态融合改进】（可见光+红外）涉及前期、中期、中后期、后期融合方式的完整配置步骤以及二次改进方案
前言
主题： YOLOv12的多模态融合改进
方式： 前期融合、中期融合、中-后期融合、后期融合。
内容： 包含融合方式详解和完整的项目包和配置步骤以及二次改进建议，开箱即用，一键运行。
一、融合方式
1.1 前期融合方法及结构图
定义： 在网络输入阶段将多模态数据直接合并，形成统一的特征表示。
实现方式： 将 RGB（3 通道）与红外（3 通道）图像直接拼接为 6 通道输入，以保留原始模态的细节信息。
结构示意图：
1.2 中期融合方法及结构图
定义： 在网络中间层（骨干网络与颈部网络之间）对多模态特征进行融合。
实现方式： 每个模态通过独立的骨干网络提取特征，融合时采用Concat操作合并特征图，送入颈部网络。
结构示意图：
1.3 中-后期融合方法及结构图
定义： 在颈部网络的上采样之后对多模态特征进行融合。
实现方式： 每个模态通过独立的骨干网络和颈部的FPN网络提取特征，融合时采用Concat操作合并特征图，送入检测头。
结构示意图：
1.4 后期融合方法及结构图
定义： 在网络输出阶段（如检测头或分类器前）对多模态特征进行融合。
实现方式： 每个模态通过独立的骨干网络和颈部网络提取特征，融合时采用Concat操作合并特征图，送入检测头。
结构示意图：
二、完整配置步骤
！！！ 私信获取的项目包就已经把相关的多模态输入、训练等改动都已经配好了，只需要新建模型yaml文件，粘贴对应的模型，进行训练即可。 项目包获取及使用教程可参考链接： 《YOLO系列模型的多模态项目》配置使用教程
在什么地方新建，n,s,m,l,x,用哪个版本按自己的需求来即可，和普通的训练步骤一致。
除了模型结构方面的改动，在yaml文件中还传入了一个通道数
ch: 6
表示传入的是双模态，6通道 ，前三个是可见光，后三个是红外。
在default.yaml中也配置了这个参数。
2.1 前期融合
# YOLOv12 🚀, AGPL-3.0 license
# YOLOv12 object detection model with P3-P5 outputs. For Usage examples see https://docs.ultralytics.com/tasks/detect
# CFG file for YOLOv12-turbo
# Parameters
ch: 6
nc: 80 # number of classes
scales: # model compound scaling constants, i.e. 'model=yolov12n.yaml' will call yolov12.yaml with scale 'n'
# [depth, width, max_channels]
n: [0.50, 0.25, 1024] # summary: 497 layers, 2,553,904 parameters, 2,553,888 gradients, 6.2 GFLOPs
s: [0.50, 0.50, 1024] # summary: 497 layers, 9,127,424 parameters, 9,127,408 gradients, 19.7 GFLOPs
m: [0.50, 1.00, 512] # summary: 533 layers, 19,670,784 parameters, 19,670,768 gradients, 60.4 GFLOPs
l: [1.00, 1.00, 512] # summary: 895 layers, 26,506,496 parameters, 26,506,480 gradients, 83.3 GFLOPs
x: [1.00, 1.50, 512] # summary: 895 layers, 59,414,176 parameters, 59,414,160 gradients, 185.9 GFLOPs
# YOLO12 backbone
backbone:
# [from, repeats, module, args]
- [-1, 1, MF, [64]] # 0
- [-1, 1, Conv, [64, 3, 2]] # 1-P1/2
- [-1, 1, Conv, [128, 3, 2, 1, 2]] # 2-P2/4
- [-1, 2, C3k2, [256, False, 0.25]]
- [-1, 1, Conv, [256, 3, 2, 1, 4]] # 4-P3/8
- [-1, 2, C3k2, [512, False, 0.25]]
- [-1, 1, Conv, [512, 3, 2]] # 6-P4/16
- [-1, 4, A2C2f, [512, True, 4]]
- [-1, 1, Conv, [1024, 3, 2]] # 8-P5/32
- [-1, 4, A2C2f, [1024, True, 1]] # 9
# YOLO12 head
head:
- [-1, 1, nn.Upsample, [None, 2, "nearest"]]
- [[-1, 7], 1, Concat, [1]] # cat backbone P4
- [-1, 2, A2C2f, [512, False, -1]] # 12
- [-1, 1, nn.Upsample, [None, 2, "nearest"]]
- [[-1, 5], 1, Concat, [1]] # cat backbone P3
- [-1, 2, A2C2f, [256, False, -1]] # 15
- [-1, 1, Conv, [256, 3, 2]]
- [[-1, 12], 1, Concat, [1]] # cat head P4
- [-1, 2, A2C2f, [512, False, -1]] # 18
- [-1, 1, Conv, [512, 3, 2]]
- [[-1, 9], 1, Concat, [1]] # cat head P5
- [-1, 2, C3k2, [1024, True]] # 21 (P5/32-large)
- [[15, 18, 21], 1, Detect, [nc]] # Detect(P3, P4, P5)
2.2 中期融合
# YOLOv12 🚀, AGPL-3.0 license
# YOLOv12 object detection model with P3-P5 outputs. For Usage examples see https://docs.ultralytics.com/tasks/detect
# CFG file for YOLOv12-turbo
# Parameters
ch: 6
nc: 80 # number of classes
scales: # model compound scaling constants, i.e. 'model=yolov12n.yaml' will call yolov12.yaml with scale 'n'
# [depth, width, max_channels]
n: [0.50, 0.25, 1024] # summary: 497 layers, 2,553,904 parameters, 2,553,888 gradients, 6.2 GFLOPs
s: [0.50, 0.50, 1024] # summary: 497 layers, 9,127,424 parameters, 9,127,408 gradients, 19.7 GFLOPs
m: [0.50, 1.00, 512] # summary: 533 layers, 19,670,784 parameters, 19,670,768 gradients, 60.4 GFLOPs
l: [1.00, 1.00, 512] # summary: 895 layers, 26,506,496 parameters, 26,506,480 gradients, 83.3 GFLOPs
x: [1.00, 1.50, 512] # summary: 895 layers, 59,414,176 parameters, 59,414,160 gradients, 185.9 GFLOPs
# YOLO12 backbone
backbone:
# [from, repeats, module, args]
- [-1, 1, IN, []] # 0
- [-1, 1, Multiin, [1]] # 1
- [-2, 1, Multiin, [2]] # 2
- [1, 1, Conv, [64, 3, 2]] # 3-P1/2
- [-1, 1, Conv, [128, 3, 2, 1, 2]] # 4-P2/4
- [-1, 2, C3k2, [256, False, 0.25]]
- [-1, 1, Conv, [256, 3, 2, 1, 4]] # 6-P3/8
- [-1, 2, C3k2, [512, False, 0.25]]
- [-1, 1, Conv, [512, 3, 2]] # 8-P4/16
- [-1, 4, A2C2f, [512, True, 4]]
- [-1, 1, Conv, [1024, 3, 2]] # 10-P5/32
- [-1, 4, A2C2f, [1024, True, 1]] # 11
- [2, 1, Conv, [64, 3, 2]] # 12-P1/2
- [-1, 1, Conv, [128, 3, 2, 1, 2]] # 13-P2/4
- [-1, 2, C3k2, [256, False, 0.25]]
- [-1, 1, Conv, [256, 3, 2, 1, 4]] # 15-P3/8
- [-1, 2, C3k2, [512, False, 0.25]]
- [-1, 1, Conv, [512, 3, 2]] # 17-P4/16
- [-1, 4, A2C2f, [512, True, 4]]
- [-1, 1, Conv, [1024, 3, 2]] # 19-P5/32
- [-1, 4, A2C2f, [1024, True, 1]] # 20
- [[7, 16], 1, Concat, [1]] # 21 cat backbone P3
- [[9, 18], 1, Concat, [1]] # 22 cat backbone P4
- [[11, 20], 1, Concat, [1]] # 23 cat backbone P5
# YOLO12 head
head:
- [-1, 1, nn.Upsample, [None, 2, "nearest"]]
- [[-1, 22], 1, Concat, [1]] # cat backbone P4
- [-1, 2, A2C2f, [512, False, -1]] # 26
- [-1, 1, nn.Upsample, [None, 2, "nearest"]]
- [[-1, 21], 1, Concat, [1]] # cat backbone P3
- [-1, 2, A2C2f, [256, False, -1]] # 29
- [-1, 1, Conv, [256, 3, 2]]
- [[-1, 26], 1, Concat, [1]] # cat head P4
- [-1, 2, A2C2f, [512, False, -1]] # 32
- [-1, 1, Conv, [512, 3, 2]]
- [[-1, 23], 1, Concat, [1]] # cat head P5
- [-1, 2, C3k2, [1024, True]] # 35 (P5/32-large)
- [[29, 32, 35], 1, Detect, [nc]] # Detect(P3, P4, P5)
2.3 中-后期融合
# YOLOv12 🚀, AGPL-3.0 license
# YOLOv12 object detection model with P3-P5 outputs. For Usage examples see https://docs.ultralytics.com/tasks/detect
# CFG file for YOLOv12-turbo
# Parameters
ch: 6
nc: 80 # number of classes
scales: # model compound scaling constants, i.e. 'model=yolov12n.yaml' will call yolov12.yaml with scale 'n'
# [depth, width, max_channels]
n: [0.50, 0.25, 1024] # summary: 497 layers, 2,553,904 parameters, 2,553,888 gradients, 6.2 GFLOPs
s: [0.50, 0.50, 1024] # summary: 497 layers, 9,127,424 parameters, 9,127,408 gradients, 19.7 GFLOPs
m: [0.50, 1.00, 512] # summary: 533 layers, 19,670,784 parameters, 19,670,768 gradients, 60.4 GFLOPs
l: [1.00, 1.00, 512] # summary: 895 layers, 26,506,496 parameters, 26,506,480 gradients, 83.3 GFLOPs
x: [1.00, 1.50, 512] # summary: 895 layers, 59,414,176 parameters, 59,414,160 gradients, 185.9 GFLOPs
# YOLO12 backbone
backbone:
# [from, repeats, module, args]
- [-1, 1, IN, []] # 0
- [-1, 1, Multiin, [1]] # 1
- [-2, 1, Multiin, [2]] # 2
- [1, 1, Conv, [64, 3, 2]] # 3-P1/2
- [-1, 1, Conv, [128, 3, 2, 1, 2]] # 4-P2/4
- [-1, 2, C3k2, [256, False, 0.25]]
- [-1, 1, Conv, [256, 3, 2, 1, 4]] # 6-P3/8
- [-1, 2, C3k2, [512, False, 0.25]]
- [-1, 1, Conv, [512, 3, 2]] # 8-P4/16
- [-1, 4, A2C2f, [512, True, 4]]
- [-1, 1, Conv, [1024, 3, 2]] # 10-P5/32
- [-1, 4, A2C2f, [1024, True, 1]] # 11
- [2, 1, Conv, [64, 3, 2]] # 12-P1/2
- [-1, 1, Conv, [128, 3, 2, 1, 2]] # 13-P2/4
- [-1, 2, C3k2, [256, False, 0.25]]
- [-1, 1, Conv, [256, 3, 2, 1, 4]] # 15-P3/8
- [-1, 2, C3k2, [512, False, 0.25]]
- [-1, 1, Conv, [512, 3, 2]] # 17-P4/16
- [-1, 4, A2C2f, [512, True, 4]]
- [-1, 1, Conv, [1024, 3, 2]] # 19-P5/32
- [-1, 4, A2C2f, [1024, True, 1]] # 20
# YOLO12 head
head:
- [11, 1, nn.Upsample, [None, 2, "nearest"]]
- [[-1, 9], 1, Concat, [1]] # cat backbone P4
- [-1, 2, A2C2f, [512, False, -1]] # 23
- [-1, 1, nn.Upsample, [None, 2, "nearest"]]
- [[-1, 7], 1, Concat, [1]] # cat backbone P3
- [-1, 2, A2C2f, [256, False, -1]] # 26
- [20, 1, nn.Upsample, [None, 2, "nearest"]]
- [[-1, 18], 1, Concat, [1]] # cat backbone P4
- [-1, 2, A2C2f, [512, False, -1]] # 29
- [-1, 1, nn.Upsample, [None, 2, "nearest"]]
- [[-1, 16], 1, Concat, [1]] # cat backbone P3
- [-1, 2, A2C2f, [256, False, -1]] # 32
- [[11, 20], 1, Concat, [1]] # cat head P5 33
- [[23, 29], 1, Concat, [1]] # cat head P5 34
- [[26, 32], 1, Concat, [1]] # cat head P5 35
- [-1, 1, Conv, [256, 3, 2]]
- [[-1, 34], 1, Concat, [1]] # cat head P4
- [-1, 2, A2C2f, [512, False, -1]] # 38
- [-1, 1, Conv, [512, 3, 2]]
- [[-1, 33], 1, Concat, [1]] # cat head P5
- [-1, 2, C3k2, [1024, True]] # 41 (P5/32-large)
- [[35, 38, 41], 1, Detect, [nc]] # Detect(P3, P4, P5)
2.4 后期融合
# YOLOv12 🚀, AGPL-3.0 license
# YOLOv12 object detection model with P3-P5 outputs. For Usage examples see https://docs.ultralytics.com/tasks/detect
# CFG file for YOLOv12-turbo
# Parameters
ch: 6
nc: 80 # number of classes
scales: # model compound scaling constants, i.e. 'model=yolov12n.yaml' will call yolov12.yaml with scale 'n'
# [depth, width, max_channels]
n: [0.50, 0.25, 1024] # summary: 497 layers, 2,553,904 parameters, 2,553,888 gradients, 6.2 GFLOPs
s: [0.50, 0.50, 1024] # summary: 497 layers, 9,127,424 parameters, 9,127,408 gradients, 19.7 GFLOPs
m: [0.50, 1.00, 512] # summary: 533 layers, 19,670,784 parameters, 19,670,768 gradients, 60.4 GFLOPs
l: [1.00, 1.00, 512] # summary: 895 layers, 26,506,496 parameters, 26,506,480 gradients, 83.3 GFLOPs
x: [1.00, 1.50, 512] # summary: 895 layers, 59,414,176 parameters, 59,414,160 gradients, 185.9 GFLOPs
# YOLO12 backbone
backbone:
# [from, repeats, module, args]
- [-1, 1, IN, []] # 0
- [-1, 1, Multiin, [1]] # 1
- [-2, 1, Multiin, [2]] # 2
- [1, 1, Conv, [64, 3, 2]] # 3-P1/2
- [-1, 1, Conv, [128, 3, 2, 1, 2]] # 4-P2/4
- [-1, 2, C3k2, [256, False, 0.25]]
- [-1, 1, Conv, [256, 3, 2, 1, 4]] # 6-P3/8
- [-1, 2, C3k2, [512, False, 0.25]]
- [-1, 1, Conv, [512, 3, 2]] # 8-P4/16
- [-1, 4, A2C2f, [512, True, 4]]
- [-1, 1, Conv, [1024, 3, 2]] # 10-P5/32
- [-1, 4, A2C2f, [1024, True, 1]] # 11
- [2, 1, Conv, [64, 3, 2]] # 12-P1/2
- [-1, 1, Conv, [128, 3, 2, 1, 2]] # 13-P2/4
- [-1, 2, C3k2, [256, False, 0.25]]
- [-1, 1, Conv, [256, 3, 2, 1, 4]] # 15-P3/8
- [-1, 2, C3k2, [512, False, 0.25]]
- [-1, 1, Conv, [512, 3, 2]] # 17-P4/16
- [-1, 4, A2C2f, [512, True, 4]]
- [-1, 1, Conv, [1024, 3, 2]] # 19-P5/32
- [-1, 4, A2C2f, [1024, True, 1]] # 20
# YOLO12 head
head:
- [11, 1, nn.Upsample, [None, 2, "nearest"]]
- [[-1, 9], 1, Concat, [1]] # cat backbone P4
- [-1, 2, A2C2f, [512, False, -1]] # 23
- [-1, 1, nn.Upsample, [None, 2, "nearest"]]
- [[-1, 7], 1, Concat, [1]] # cat backbone P3
- [-1, 2, A2C2f, [256, False, -1]] # 26
- [-1, 1, Conv, [256, 3, 2]]
- [[-1, 23], 1, Concat, [1]] # cat head P4
- [-1, 2, A2C2f, [512, False, -1]] # 29
- [-1, 1, Conv, [512, 3, 2]]
- [[-1, 11], 1, Concat, [1]] # cat head P5
- [-1, 2, C3k2, [1024, True]] # 32 (P5/32-large)
- [20, 1, nn.Upsample, [None, 2, "nearest"]]
- [[-1, 18], 1, Concat, [1]] # cat backbone P4
- [-1, 2, A2C2f, [512, False, -1]] # 35
- [-1, 1, nn.Upsample, [None, 2, "nearest"]]
- [[-1, 16], 1, Concat, [1]] # cat backbone P3
- [-1, 2, A2C2f, [256, False, -1]] # 38
- [-1, 1, Conv, [256, 3, 2]]
- [[-1, 35], 1, Concat, [1]] # cat head P4
- [-1, 2, A2C2f, [512, False, -1]] # 41
- [-1, 1, Conv, [512, 3, 2]]
- [[-1, 20], 1, Concat, [1]] # cat head P5
- [-1, 2, C3k2, [1024, True]] # 44 (P5/32-large)
- [[26, 38], 1, Concat, [1]] # cat head P5 45
- [[29, 41], 1, Concat, [1]] # cat head P5 46
- [[32, 44], 1, Concat, [1]] # cat head P5 47
- [[45, 46, 47], 1, Detect, [nc]] # Detect(P3, P4, P5)
三、成功运行结果
前期融合结果： 可以看到输入的通道数为6，表明可见光图像和红外图像均输入到了模型中进行融合训练。
YOLOv12-earlyfusion summary: 479 layers, 2,540,635 parameters, 2,540,619 gradients, 6.8 GFLOPs
from n params module arguments
0 -1 1 472 ultralytics.nn.AddModules.multimodal.MF [6, 16]
1 -1 1 2336 ultralytics.nn.modules.conv.Conv [16, 16, 3, 2]
2 -1 1 2368 ultralytics.nn.modules.conv.Conv [16, 32, 3, 2, 1, 2]
3 -1 1 6640 ultralytics.nn.modules.block.C3k2 [32, 64, 1, False, 0.25]
4 -1 1 9344 ultralytics.nn.modules.conv.Conv [64, 64, 3, 2, 1, 4]
5 -1 1 26080 ultralytics.nn.modules.block.C3k2 [64, 128, 1, False, 0.25]
6 -1 1 147712 ultralytics.nn.modules.conv.Conv [128, 128, 3, 2]
7 -1 2 180864 ultralytics.nn.AddModules.A2C2f.A2C2f [128, 128, 2, True, 4]
8 -1 1 295424 ultralytics.nn.modules.conv.Conv [128, 256, 3, 2]
9 -1 2 689408 ultralytics.nn.AddModules.A2C2f.A2C2f [256, 256, 2, True, 1]
10 -1 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest']
11 [-1, 7] 1 0 ultralytics.nn.modules.conv.Concat [1]
12 -1 1 86912 ultralytics.nn.AddModules.A2C2f.A2C2f [384, 128, 1, False, -1]
13 -1 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest']
14 [-1, 5] 1 0 ultralytics.nn.modules.conv.Concat [1]
15 -1 1 24000 ultralytics.nn.AddModules.A2C2f.A2C2f [256, 64, 1, False, -1]
16 -1 1 36992 ultralytics.nn.modules.conv.Conv [64, 64, 3, 2]
17 [-1, 12] 1 0 ultralytics.nn.modules.conv.Concat [1]
18 -1 1 74624 ultralytics.nn.AddModules.A2C2f.A2C2f [192, 128, 1, False, -1]
19 -1 1 147712 ultralytics.nn.modules.conv.Conv [128, 128, 3, 2]
20 [-1, 9] 1 0 ultralytics.nn.modules.conv.Concat [1]
21 -1 1 378880 ultralytics.nn.modules.block.C3k2 [384, 256, 1, True]
22 [15, 18, 21] 1 430867 ultralytics.nn.modules.head.Detect [1, [64, 128, 256]]
YOLOv12-earlyfusion summary: 479 layers, 2,540,635 parameters, 2,540,619 gradients, 6.8 GFLOPs
中期融合结果：
YOLOv12-midfusion summary: 713 layers, 3,990,803 parameters, 3,990,787 gradients, 9.1 GFLOPs
from n params module arguments
0 -1 1 0 ultralytics.nn.AddModules.multimodal.IN []
1 -1 1 0 ultralytics.nn.AddModules.multimodal.Multiin [1]
2 -2 1 0 ultralytics.nn.AddModules.multimodal.Multiin [2]
3 1 1 464 ultralytics.nn.modules.conv.Conv [3, 16, 3, 2]
4 -1 1 2368 ultralytics.nn.modules.conv.Conv [16, 32, 3, 2, 1, 2]
5 -1 1 6640 ultralytics.nn.modules.block.C3k2 [32, 64, 1, False, 0.25]
6 -1 1 9344 ultralytics.nn.modules.conv.Conv [64, 64, 3, 2, 1, 4]
7 -1 1 26080 ultralytics.nn.modules.block.C3k2 [64, 128, 1, False, 0.25]
8 -1 1 147712 ultralytics.nn.modules.conv.Conv [128, 128, 3, 2]
9 -1 2 180864 ultralytics.nn.AddModules.A2C2f.A2C2f [128, 128, 2, True, 4]
10 -1 1 295424 ultralytics.nn.modules.conv.Conv [128, 256, 3, 2]
11 -1 2 689408 ultralytics.nn.AddModules.A2C2f.A2C2f [256, 256, 2, True, 1]
12 2 1 464 ultralytics.nn.modules.conv.Conv [3, 16, 3, 2]
13 -1 1 2368 ultralytics.nn.modules.conv.Conv [16, 32, 3, 2, 1, 2]
14 -1 1 6640 ultralytics.nn.modules.block.C3k2 [32, 64, 1, False, 0.25]
15 -1 1 9344 ultralytics.nn.modules.conv.Conv [64, 64, 3, 2, 1, 4]
16 -1 1 26080 ultralytics.nn.modules.block.C3k2 [64, 128, 1, False, 0.25]
17 -1 1 147712 ultralytics.nn.modules.conv.Conv [128, 128, 3, 2]
18 -1 2 180864 ultralytics.nn.AddModules.A2C2f.A2C2f [128, 128, 2, True, 4]
19 -1 1 295424 ultralytics.nn.modules.conv.Conv [128, 256, 3, 2]
20 -1 2 689408 ultralytics.nn.AddModules.A2C2f.A2C2f [256, 256, 2, True, 1]
21 [7, 16] 1 0 ultralytics.nn.modules.conv.Concat [1]
22 [9, 18] 1 0 ultralytics.nn.modules.conv.Concat [1]
23 [11, 20] 1 0 ultralytics.nn.modules.conv.Concat [1]
24 -1 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest']
25 [-1, 22] 1 0 ultralytics.nn.modules.conv.Concat [1]
26 -1 1 111488 ultralytics.nn.AddModules.A2C2f.A2C2f [768, 128, 1, False, -1]
27 -1 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest']
28 [-1, 21] 1 0 ultralytics.nn.modules.conv.Concat [1]
29 -1 1 28096 ultralytics.nn.AddModules.A2C2f.A2C2f [384, 64, 1, False, -1]
30 -1 1 36992 ultralytics.nn.modules.conv.Conv [64, 64, 3, 2]
31 [-1, 26] 1 0 ultralytics.nn.modules.conv.Concat [1]
32 -1 1 74624 ultralytics.nn.AddModules.A2C2f.A2C2f [192, 128, 1, False, -1]
33 -1 1 147712 ultralytics.nn.modules.conv.Conv [128, 128, 3, 2]
34 [-1, 23] 1 0 ultralytics.nn.modules.conv.Concat [1]
35 -1 1 444416 ultralytics.nn.modules.block.C3k2 [640, 256, 1, True]
36 [29, 32, 35] 1 430867 ultralytics.nn.modules.head.Detect [1, [64, 128, 256]]
YOLOv12-midfusion summary: 713 layers, 3,990,803 parameters, 3,990,787 gradients, 9.1 GFLOPs
中-后期融合结果：
YOLOv12-mid-to-late-fusion summary: 783 layers, 4,232,467 parameters, 4,232,451 gradients, 10.6 GFLOPs
from n params module arguments
0 -1 1 0 ultralytics.nn.AddModules.multimodal.IN []
1 -1 1 0 ultralytics.nn.AddModules.multimodal.Multiin [1]
2 -2 1 0 ultralytics.nn.AddModules.multimodal.Multiin [2]
3 1 1 464 ultralytics.nn.modules.conv.Conv [3, 16, 3, 2]
4 -1 1 2368 ultralytics.nn.modules.conv.Conv [16, 32, 3, 2, 1, 2]
5 -1 1 6640 ultralytics.nn.modules.block.C3k2 [32, 64, 1, False, 0.25]
6 -1 1 9344 ultralytics.nn.modules.conv.Conv [64, 64, 3, 2, 1, 4]
7 -1 1 26080 ultralytics.nn.modules.block.C3k2 [64, 128, 1, False, 0.25]
8 -1 1 147712 ultralytics.nn.modules.conv.Conv [128, 128, 3, 2]
9 -1 2 180864 ultralytics.nn.AddModules.A2C2f.A2C2f [128, 128, 2, True, 4]
10 -1 1 295424 ultralytics.nn.modules.conv.Conv [128, 256, 3, 2]
11 -1 2 689408 ultralytics.nn.AddModules.A2C2f.A2C2f [256, 256, 2, True, 1]
12 2 1 464 ultralytics.nn.modules.conv.Conv [3, 16, 3, 2]
13 -1 1 2368 ultralytics.nn.modules.conv.Conv [16, 32, 3, 2, 1, 2]
14 -1 1 6640 ultralytics.nn.modules.block.C3k2 [32, 64, 1, False, 0.25]
15 -1 1 9344 ultralytics.nn.modules.conv.Conv [64, 64, 3, 2, 1, 4]
16 -1 1 26080 ultralytics.nn.modules.block.C3k2 [64, 128, 1, False, 0.25]
17 -1 1 147712 ultralytics.nn.modules.conv.Conv [128, 128, 3, 2]
18 -1 2 180864 ultralytics.nn.AddModules.A2C2f.A2C2f [128, 128, 2, True, 4]
19 -1 1 295424 ultralytics.nn.modules.conv.Conv [128, 256, 3, 2]
20 -1 2 689408 ultralytics.nn.AddModules.A2C2f.A2C2f [256, 256, 2, True, 1]
21 11 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest']
22 [-1, 9] 1 0 ultralytics.nn.modules.conv.Concat [1]
23 -1 1 86912 ultralytics.nn.AddModules.A2C2f.A2C2f [384, 128, 1, False, -1]
24 -1 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest']
25 [-1, 7] 1 0 ultralytics.nn.modules.conv.Concat [1]
26 -1 1 24000 ultralytics.nn.AddModules.A2C2f.A2C2f [256, 64, 1, False, -1]
27 20 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest']
28 [-1, 18] 1 0 ultralytics.nn.modules.conv.Concat [1]
29 -1 1 86912 ultralytics.nn.AddModules.A2C2f.A2C2f [384, 128, 1, False, -1]
30 -1 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest']
31 [-1, 16] 1 0 ultralytics.nn.modules.conv.Concat [1]
32 -1 1 24000 ultralytics.nn.AddModules.A2C2f.A2C2f [256, 64, 1, False, -1]
33 [11, 20] 1 0 ultralytics.nn.modules.conv.Concat [1]
34 [23, 29] 1 0 ultralytics.nn.modules.conv.Concat [1]
35 [26, 32] 1 0 ultralytics.nn.modules.conv.Concat [1]
36 -1 1 73856 ultralytics.nn.modules.conv.Conv [128, 64, 3, 2]
37 [-1, 34] 1 0 ultralytics.nn.modules.conv.Concat [1]
38 -1 1 82816 ultralytics.nn.AddModules.A2C2f.A2C2f [320, 128, 1, False, -1]
39 -1 1 147712 ultralytics.nn.modules.conv.Conv [128, 128, 3, 2]
40 [-1, 33] 1 0 ultralytics.nn.modules.conv.Concat [1]
41 -1 1 444416 ultralytics.nn.modules.block.C3k2 [640, 256, 1, True]
42 [35, 38, 41] 1 545235 ultralytics.nn.modules.head.Detect [1, [128, 128, 256]]
YOLOv12-mid-to-late-fusion summary: 783 layers, 4,232,467 parameters, 4,232,451 gradients, 10.6 GFLOPs
后期融合结果：
YOLOv12-latefusion summary: 857 layers, 5,034,643 parameters, 5,034,627 gradients, 11.6 GFLOPs
from n params module arguments
0 -1 1 0 ultralytics.nn.AddModules.multimodal.IN []
1 -1 1 0 ultralytics.nn.AddModules.multimodal.Multiin [1]
2 -2 1 0 ultralytics.nn.AddModules.multimodal.Multiin [2]
3 1 1 464 ultralytics.nn.modules.conv.Conv [3, 16, 3, 2]
4 -1 1 2368 ultralytics.nn.modules.conv.Conv [16, 32, 3, 2, 1, 2]
5 -1 1 6640 ultralytics.nn.modules.block.C3k2 [32, 64, 1, False, 0.25]
6 -1 1 9344 ultralytics.nn.modules.conv.Conv [64, 64, 3, 2, 1, 4]
7 -1 1 26080 ultralytics.nn.modules.block.C3k2 [64, 128, 1, False, 0.25]
8 -1 1 147712 ultralytics.nn.modules.conv.Conv [128, 128, 3, 2]
9 -1 2 180864 ultralytics.nn.AddModules.A2C2f.A2C2f [128, 128, 2, True, 4]
10 -1 1 295424 ultralytics.nn.modules.conv.Conv [128, 256, 3, 2]
11 -1 2 689408 ultralytics.nn.AddModules.A2C2f.A2C2f [256, 256, 2, True, 1]
12 2 1 464 ultralytics.nn.modules.conv.Conv [3, 16, 3, 2]
13 -1 1 2368 ultralytics.nn.modules.conv.Conv [16, 32, 3, 2, 1, 2]
14 -1 1 6640 ultralytics.nn.modules.block.C3k2 [32, 64, 1, False, 0.25]
15 -1 1 9344 ultralytics.nn.modules.conv.Conv [64, 64, 3, 2, 1, 4]
16 -1 1 26080 ultralytics.nn.modules.block.C3k2 [64, 128, 1, False, 0.25]
17 -1 1 147712 ultralytics.nn.modules.conv.Conv [128, 128, 3, 2]
18 -1 2 180864 ultralytics.nn.AddModules.A2C2f.A2C2f [128, 128, 2, True, 4]
19 -1 1 295424 ultralytics.nn.modules.conv.Conv [128, 256, 3, 2]
20 -1 2 689408 ultralytics.nn.AddModules.A2C2f.A2C2f [256, 256, 2, True, 1]
21 11 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest']
22 [-1, 9] 1 0 ultralytics.nn.modules.conv.Concat [1]
23 -1 1 86912 ultralytics.nn.AddModules.A2C2f.A2C2f [384, 128, 1, False, -1]
24 -1 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest']
25 [-1, 7] 1 0 ultralytics.nn.modules.conv.Concat [1]
26 -1 1 24000 ultralytics.nn.AddModules.A2C2f.A2C2f [256, 64, 1, False, -1]
27 -1 1 36992 ultralytics.nn.modules.conv.Conv [64, 64, 3, 2]
28 [-1, 23] 1 0 ultralytics.nn.modules.conv.Concat [1]
29 -1 1 74624 ultralytics.nn.AddModules.A2C2f.A2C2f [192, 128, 1, False, -1]
30 -1 1 147712 ultralytics.nn.modules.conv.Conv [128, 128, 3, 2]
31 [-1, 11] 1 0 ultralytics.nn.modules.conv.Concat [1]
32 -1 1 378880 ultralytics.nn.modules.block.C3k2 [384, 256, 1, True]
33 20 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest']
34 [-1, 18] 1 0 ultralytics.nn.modules.conv.Concat [1]
35 -1 1 86912 ultralytics.nn.AddModules.A2C2f.A2C2f [384, 128, 1, False, -1]
36 -1 1 0 torch.nn.modules.upsampling.Upsample [None, 2, 'nearest']
37 [-1, 16] 1 0 ultralytics.nn.modules.conv.Concat [1]
38 -1 1 24000 ultralytics.nn.AddModules.A2C2f.A2C2f [256, 64, 1, False, -1]
39 -1 1 36992 ultralytics.nn.modules.conv.Conv [64, 64, 3, 2]
40 [-1, 35] 1 0 ultralytics.nn.modules.conv.Concat [1]
41 -1 1 74624 ultralytics.nn.AddModules.A2C2f.A2C2f [192, 128, 1, False, -1]
42 -1 1 147712 ultralytics.nn.modules.conv.Conv [128, 128, 3, 2]
43 [-1, 20] 1 0 ultralytics.nn.modules.conv.Concat [1]
44 -1 1 378880 ultralytics.nn.modules.block.C3k2 [384, 256, 1, True]
45 [26, 38] 1 0 ultralytics.nn.modules.conv.Concat [1]
46 [29, 41] 1 0 ultralytics.nn.modules.conv.Concat [1]
47 [32, 44] 1 0 ultralytics.nn.modules.conv.Concat [1]
48 [45, 46, 47] 1 819795 ultralytics.nn.modules.head.Detect [1, [128, 256, 512]]
YOLOv12-latefusion summary: 857 layers, 5,034,643 parameters, 5,034,627 gradients, 11.6 GFLOPs
四、二次改进方案
-
多模态模型的二次改进和普通模型的改进一致，主要涉及到A2C2f、颈部结构、上采样、下采样等，可以增加或替换成其它模块，可以换成其它的颈部结构在进行融合。若有需要可查看主页的模块改进专栏。
-
两个骨干中均可以再次添加其它模块，需要注意的是融合的时候层数要对应上，即两层的特征图大小要一致。