Network Design of Yolov5

[flixmk]

❔Question

I have a question about the network design of yolov5. When writing a somewhat scientific essay about my work with Yolov5 I need to describe the network architecture. Has anything changed in this regard from yolov3? And is the paper oy Alexey for yolov4 still useable for this network?

And I have a second question. Does Yolo have Pooling Layers? I have heard that it doesnt, but when looking at the Darknet-53 backbone it seems to have a Avgpool layer at the end? Am I missing something?

Additional context

[glenn-jocher]

@Kraufel model architecture is defined by yaml files in the models/directory. i.e.:

yolov5/models/yolov5s.yaml

Lines 1 to 48 in 3e25f1e

	# parameters
	nc: 80 # number of classes
	depth_multiple: 0.33 # model depth multiple
	width_multiple: 0.50 # layer channel multiple
	# anchors
	anchors:
	- [10,13, 16,30, 33,23] # P3/8
	- [30,61, 62,45, 59,119] # P4/16
	- [116,90, 156,198, 373,326] # P5/32
	# YOLOv5 backbone
	backbone:
	# [from, number, module, args]
	[[-1, 1, Focus, [64, 3]], # 0-P1/2
	[-1, 1, Conv, [128, 3, 2]], # 1-P2/4
	[-1, 3, C3, [128]],
	[-1, 1, Conv, [256, 3, 2]], # 3-P3/8
	[-1, 9, C3, [256]],
	[-1, 1, Conv, [512, 3, 2]], # 5-P4/16
	[-1, 9, C3, [512]],
	[-1, 1, Conv, [1024, 3, 2]], # 7-P5/32
	[-1, 1, SPP, [1024, [5, 9, 13]]],
	[-1, 3, C3, [1024, False]], # 9
	]
	# YOLOv5 head
	head:
	[[-1, 1, Conv, [512, 1, 1]],
	[-1, 1, nn.Upsample, [None, 2, 'nearest']],
	[[-1, 6], 1, Concat, [1]], # cat backbone P4
	[-1, 3, C3, [512, False]], # 13
	[-1, 1, Conv, [256, 1, 1]],
	[-1, 1, nn.Upsample, [None, 2, 'nearest']],
	[[-1, 4], 1, Concat, [1]], # cat backbone P3
	[-1, 3, C3, [256, False]], # 17 (P3/8-small)
	[-1, 1, Conv, [256, 3, 2]],
	[[-1, 14], 1, Concat, [1]], # cat head P4
	[-1, 3, C3, [512, False]], # 20 (P4/16-medium)
	[-1, 1, Conv, [512, 3, 2]],
	[[-1, 10], 1, Concat, [1]], # cat head P5
	[-1, 3, C3, [1024, False]], # 23 (P5/32-large)
	[[17, 20, 23], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5)
	]

[flixmk]

@glenn-jocher Thank you. But what kind of backbone, head and neck is used? Backbone doesnt seem to be Darknet-53.

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.

[zhiqwang]

Hi @Kraufel, I've refactored the backbone and neck from the yaml config architecture in here, maybe you could refer to these.

BTW, here is an diagram overview of YOLOv5 release 4.0 .

[github-actions]

This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions.