目录
1.ASFF介绍
2.ASFF加入Yolov5提升检测精度
2.1 ASFF加入common.py中：
2.2 ASFF加入yolo.py中：
2.3 修改yolov5s_asff.yaml
2.4 与cbam结合 进一步提升检测精度
1.ASFF介绍
Learning SpatialFusionfor Single-Shot Object Detection
论文地址：https://arxiv.org/pdf/1911.09516v2.pdf
多尺度特征特别是特征金字塔FPN是解决目标检测中跨尺度目标的最常用有效的解决方法，但是不同特征尺度中存在的不一致性限制了（基于特征金字塔的）single-shot检测器的性能。本文提出一种特征金字塔融合方法ASFF，它自动学习去抑制不同尺度特征在融合时空间上可能存在不一致；
2.ASFF加入Yolov5提升检测精度
2.1 ASFF加入common.py中：
class ASFFV5(nn.Module):def __init__(self, level, multiplier=1, rfb=False, vis=False, act_cfg=True):"""ASFF version for YoloV5 .different than YoloV3multiplier should be 1, 0.5which means, the channel of ASFF can be512, 256, 128 -> multiplier=1256, 128, 64 -> multiplier=0.5For even smaller, you need change code manually."""super(ASFFV5, self).__init__()self.level = levelself.dim = [int(1024 * multiplier), int(512 * multiplier),int(256 * multiplier)]# print(self.dim)self.inter_dim = self.dim[self.level]if level == 0:self.stride_level_1 = Conv(int(512 * multiplier), self.inter_dim, 3, 2)self.stride_level_2 = Conv(int(256 * multiplier), self.inter_dim, 3, 2)self.expand = Conv(self.inter_dim, int(1024 * multiplier), 3, 1)elif level == 1:self.compress_level_0 = Conv(int(1024 * multiplier), self.inter_dim, 1, 1)self.stride_level_2 = Conv(int(256 * multiplier), self.inter_dim, 3, 2)self.expand = Conv(self.inter_dim, int(512 * multiplier), 3, 1)elif level == 2:self.compress_level_0 = Conv(int(1024 * multiplier), self.inter_dim, 1, 1)self.compress_level_1 = Conv(int(512 * multiplier), self.inter_dim, 1, 1)self.expand = Conv(self.inter_dim, int(256 * multiplier), 3, 1)# when adding rfb, we use half number of channels to save memorycompress_c = 8 if rfb else 16self.weight_level_0 = Conv(self.inter_dim, compress_c, 1, 1)self.weight_level_1 = Conv(self.inter_dim, compress_c, 1, 1)self.weight_level_2 = Conv(self.inter_dim, compress_c, 1, 1)self.weight_levels = Conv(compress_c * 3, 3, 1, 1)self.vis = visdef forward(self, x):# l,m,s"""# 128, 256, 512512, 256, 128from small -> large"""x_level_0 = x[2]# lx_level_1 = x[1]# mx_level_2 = x[0]# s# print('x_level_0: ', x_level_0.shape)# print('x_level_1: ', x_level_1.shape)# print('x_level_2: ', x_level_2.shape)if self.level == 0:level_0_resized = x_level_0level_1_resized = self.stride_level_1(x_level_1)level_2_downsampled_inter = F.max_pool2d(x_level_2, 3, stride=2, padding=1)level_2_resized = self.stride_level_2(level_2_downsampled_inter)elif self.level == 1:level_0_compressed = self.compress_level_0(x_level_0)level_0_resized = F.interpolate(level_0_compressed, scale_factor=2, mode='nearest')level_1_resized = x_level_1level_2_resized = self.stride_level_2(x_level_2)elif self.level == 2:level_0_compressed = self.compress_level_0(x_level_0)level_0_resized = F.interpolate(level_0_compressed, scale_factor=4, mode='nearest')x_level_1_compressed = self.compress_level_1(x_level_1)level_1_resized = F.interpolate(x_level_1_compressed, scale_factor=2, mode='nearest')level_2_resized = x_level_2# print('level: {}, l1_resized: {}, l2_resized: {}'.format(self.level,#level_1_resized.shape, level_2_resized.shape))level_0_weight_v = self.weight_level_0(level_0_resized)level_1_weight_v = self.weight_level_1(level_1_resized)level_2_weight_v = self.weight_level_2(level_2_resized)# print('level_0_weight_v: ', level_0_weight_v.shape)# print('level_1_weight_v: ', level_1_weight_v.shape)# print('level_2_weight_v: ', level_2_weight_v.shape)levels_weight_v = torch.cat((level_0_weight_v, level_1_weight_v, level_2_weight_v), 1)levels_weight = self.weight_levels(levels_weight_v)levels_weight = F.softmax(levels_weight, dim=1)fused_out_reduced = level_0_resized * levels_weight[:, 0:1, :, :] + \level_1_resized * levels_weight[:, 1:2, :, :] + \level_2_resized * levels_weight[:, 2:, :, :]out = self.expand(fused_out_reduced)if self.vis:return out, levels_weight, fused_out_reduced.sum(dim=1)else:return out# ------------------------------------asff -----end--------------------------------2.2 ASFF加入yolo.py中：
class ASFF_Detect(nn.Module):# add ASFFV5 layer and Rfbstride = None# strides computed during buildonnx_dynamic = False# ONNX export parameterexport = False# export modedef __init__(self, nc=80, anchors=(), ch=(), multiplier=0.5, rfb=False, inplace=True):# detection layersuper().__init__()self.nc = nc# number of classesself.no = nc + 5# number of outputs per anchorself.nl = len(anchors)# number of detection layersself.na = len(anchors[0]) // 2# number of anchorsself.grid = [torch.zeros(1)] * self.nl# init gridself.l0_fusion = ASFFV5(level=0, multiplier=multiplier, rfb=rfb)self.l1_fusion = ASFFV5(level=1, multiplier=multiplier, rfb=rfb)self.l2_fusion = ASFFV5(level=2, multiplier=multiplier, rfb=rfb)self.anchor_grid = [torch.zeros(1)] * self.nl# init anchor gridself.register_buffer('anchors', torch.tensor(anchors).float().view(self.nl, -1, 2))# shape(nl,na,2)self.m = nn.ModuleList(nn.Conv2d(x, self.no * self.na, 1) for x in ch)# output convself.inplace = inplace# use in-place ops (e.g. slice assignment)def forward(self, x):z = []# inference outputresult = []result.append(self.l2_fusion(x))result.append(self.l1_fusion(x))result.append(self.l0_fusion(x))x = resultfor i in range(self.nl):x[i] = self.m[i](x[i])# convbs, _, ny, nx = x[i].shape# x(bs,255,20,20) to x(bs,3,20,20,85)x[i] = x[i].view(bs, self.na, self.no, ny, nx).permute(0, 1, 3, 4, 2).contiguous()if not self.training:# inferenceif self.onnx_dynamic or self.grid[i].shape[2:4] != x[i].shape[2:4]:self.grid[i], self.anchor_grid[i] = self._make_grid(nx, ny, i)y = x[i].sigmoid()if self.inplace:y[..., 0:2] = (y[..., 0:2] * 2 + self.grid[i]) * self.stride[i]# xyy[..., 2:4] = (y[..., 2:4] * 2) ** 2 * self.anchor_grid[i]# whelse:# for YOLOv5 on AWS Inferentia https://github.com/ultralytics/yolov5/pull/2953xy, wh, conf = y.split((2, 2, self.nc + 1), 4)# y.tensor_split((2, 4, 5), 4)# torch 1.8.0xy = (xy * 2 + self.grid[i]) * self.stride[i]# xywh = (wh * 2) ** 2 * self.anchor_grid[i]# why = torch.cat((xy, wh, conf), 4)z.append(y.view(bs, -1, self.no))return x if self.training else (torch.cat(z, 1),) if self.export else (torch.cat(z, 1), x)def _make_grid(self, nx=20, ny=20, i=0, torch_1_10=check_version(torch.__version__, '1.10.0')):d = self.anchors[i].devicet = self.anchors[i].dtypeshape = 1, self.na, ny, nx, 2# grid shapey, x = torch.arange(ny, device=d, dtype=t), torch.arange(nx, device=d, dtype=t)if torch_1_10:# torch>=1.10.0 meshgrid workaround for torch>=0.7 compatibilityyv, xv = torch.meshgrid(y, x, indexing='ij')else:yv, xv = torch.meshgrid(y, x)grid = torch.stack((xv, yv), 2).expand(shape) - 0.5# add grid offset, i.e. y = 2.0 * x - 0.5anchor_grid = (self.anchors[i] * self.stride[i]).view((1, self.na, 1, 1, 2)).expand(shape)# print(anchor_grid)return grid, anchor_gridclass DetectionModel(BaseModel):下加入 （PS:建议直接搜索Detect关键词）
m = self.model[-1]# Detect()if isinstance(m, (Detect, Segment,ASFF_Detect)):def parse_model(d, ch): # model_dict, input_channels(3)
# TODO: channel, gw, gdelif m in {Detect, Segment,ASFF_Detect}:args.append([ch[x] for x in f])class BaseModel(nn.Module):
def _apply(self, fn):# Apply to(), cpu(), cuda(), half() to model tensors that are not parameters or registered buffersself = super()._apply(fn)m = self.model[-1]# Detect()if isinstance(m, (Detect, Segment,ASFF_Detect)):2.3 修改yolov5s_asff.yaml
# YOLOv5by Ultralytics, GPL-3.0 license# Parametersnc: 1# number of classesdepth_multiple: 0.33# model depth multiplewidth_multiple: 0.50# layer channel multipleanchors:- [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 v6.0 backbonebackbone:# [from, number, module, args][[-1, 1, Conv, [64, 6, 2, 2]],# 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, 6, 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, 3, C3, [1024]], [-1, 1, SPPF, [1024, 5]],# 9]# YOLOv5 v6.0 headhead:[[-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, ASFF_Detect, [nc, anchors]],# Detect(P3, P4, P5)]2.4 与cbam结合 进一步提升检测精度
cbam介绍：https://blog.csdn.net/m0_63774211/article/details/129611391
# Parametersnc: 1 # number of classesdepth_multiple: 0.67# model depth multiplewidth_multiple: 0.75# layer channel multiple# anchorsanchors:- [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 v6.0 backbonebackbone:# [from, number, module, args][[-1, 1, Conv, [64, 6, 2, 2]],# 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, 6, 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, 3, C3, [1024]], [-1, 1, CBAM, [1024]], #9 [-1, 1, SPPF, [1024, 5]],#10]# YOLOv5 v6.0 headhead:[[-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]],# 14 [-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]],# 18 (P3/8-small) [-1, 1, CBAM, [256]], #19 [-1, 1, Conv, [256, 3, 2]], [[-1, 14], 1, Concat, [1]],# cat head P4 [-1, 3, C3, [512, False]],# 22 (P4/16-medium) [-1, 1, CBAM, [512]], [-1, 1, Conv, [512, 3, 2]], [[-1, 10], 1, Concat, [1]],# cat head P5 [-1, 3, C3, [1024, False]],# 25 (P5/32-large) [-1, 1, CBAM, [1024]], [[19, 23, 27], 1, ASFF_Detect, [nc, anchors]],# Detect(P3, P4, P5)]