diff options
Diffstat (limited to 'r_facelib/detection/yolov5face')
| -rw-r--r-- | r_facelib/detection/yolov5face/__init__.py | 0 | ||||
| -rw-r--r-- | r_facelib/detection/yolov5face/face_detector.py | 141 | ||||
| -rw-r--r-- | r_facelib/detection/yolov5face/models/__init__.py | 0 | ||||
| -rw-r--r-- | r_facelib/detection/yolov5face/models/common.py | 299 | ||||
| -rw-r--r-- | r_facelib/detection/yolov5face/models/experimental.py | 45 | ||||
| -rw-r--r-- | r_facelib/detection/yolov5face/models/yolo.py | 235 | ||||
| -rw-r--r-- | r_facelib/detection/yolov5face/models/yolov5l.yaml | 47 | ||||
| -rw-r--r-- | r_facelib/detection/yolov5face/models/yolov5n.yaml | 45 | ||||
| -rw-r--r-- | r_facelib/detection/yolov5face/utils/__init__.py | 0 | ||||
| -rw-r--r-- | r_facelib/detection/yolov5face/utils/autoanchor.py | 12 | ||||
| -rw-r--r-- | r_facelib/detection/yolov5face/utils/datasets.py | 35 | ||||
| -rw-r--r-- | r_facelib/detection/yolov5face/utils/extract_ckpt.py | 5 | ||||
| -rw-r--r-- | r_facelib/detection/yolov5face/utils/general.py | 271 | ||||
| -rw-r--r-- | r_facelib/detection/yolov5face/utils/torch_utils.py | 40 |
14 files changed, 1175 insertions, 0 deletions
diff --git a/r_facelib/detection/yolov5face/__init__.py b/r_facelib/detection/yolov5face/__init__.py new file mode 100644 index 0000000..e69de29 --- /dev/null +++ b/r_facelib/detection/yolov5face/__init__.py diff --git a/r_facelib/detection/yolov5face/face_detector.py b/r_facelib/detection/yolov5face/face_detector.py new file mode 100644 index 0000000..ca6d8e3 --- /dev/null +++ b/r_facelib/detection/yolov5face/face_detector.py @@ -0,0 +1,141 @@ +import copy
+from pathlib import Path
+
+import cv2
+import numpy as np
+import torch
+from torch import torch_version
+
+from r_facelib.detection.yolov5face.models.common import Conv
+from r_facelib.detection.yolov5face.models.yolo import Model
+from r_facelib.detection.yolov5face.utils.datasets import letterbox
+from r_facelib.detection.yolov5face.utils.general import (
+ check_img_size,
+ non_max_suppression_face,
+ scale_coords,
+ scale_coords_landmarks,
+)
+
+print(f"Torch version: {torch.__version__}")
+IS_HIGH_VERSION = torch_version.__version__ >= "1.9.0"
+
+def isListempty(inList):
+ if isinstance(inList, list): # Is a list
+ return all(map(isListempty, inList))
+ return False # Not a list
+
+class YoloDetector:
+ def __init__(
+ self,
+ config_name,
+ min_face=10,
+ target_size=None,
+ device='cuda',
+ ):
+ """
+ config_name: name of .yaml config with network configuration from models/ folder.
+ min_face : minimal face size in pixels.
+ target_size : target size of smaller image axis (choose lower for faster work). e.g. 480, 720, 1080.
+ None for original resolution.
+ """
+ self._class_path = Path(__file__).parent.absolute()
+ self.target_size = target_size
+ self.min_face = min_face
+ self.detector = Model(cfg=config_name)
+ self.device = device
+
+
+ def _preprocess(self, imgs):
+ """
+ Preprocessing image before passing through the network. Resize and conversion to torch tensor.
+ """
+ pp_imgs = []
+ for img in imgs:
+ h0, w0 = img.shape[:2] # orig hw
+ if self.target_size:
+ r = self.target_size / min(h0, w0) # resize image to img_size
+ if r < 1:
+ img = cv2.resize(img, (int(w0 * r), int(h0 * r)), interpolation=cv2.INTER_LINEAR)
+
+ imgsz = check_img_size(max(img.shape[:2]), s=self.detector.stride.max()) # check img_size
+ img = letterbox(img, new_shape=imgsz)[0]
+ pp_imgs.append(img)
+ pp_imgs = np.array(pp_imgs)
+ pp_imgs = pp_imgs.transpose(0, 3, 1, 2)
+ pp_imgs = torch.from_numpy(pp_imgs).to(self.device)
+ pp_imgs = pp_imgs.float() # uint8 to fp16/32
+ return pp_imgs / 255.0 # 0 - 255 to 0.0 - 1.0
+
+ def _postprocess(self, imgs, origimgs, pred, conf_thres, iou_thres):
+ """
+ Postprocessing of raw pytorch model output.
+ Returns:
+ bboxes: list of arrays with 4 coordinates of bounding boxes with format x1,y1,x2,y2.
+ points: list of arrays with coordinates of 5 facial keypoints (eyes, nose, lips corners).
+ """
+ bboxes = [[] for _ in range(len(origimgs))]
+ landmarks = [[] for _ in range(len(origimgs))]
+
+ pred = non_max_suppression_face(pred, conf_thres, iou_thres)
+
+ for image_id, origimg in enumerate(origimgs):
+ img_shape = origimg.shape
+ image_height, image_width = img_shape[:2]
+ gn = torch.tensor(img_shape)[[1, 0, 1, 0]] # normalization gain whwh
+ gn_lks = torch.tensor(img_shape)[[1, 0, 1, 0, 1, 0, 1, 0, 1, 0]] # normalization gain landmarks
+ det = pred[image_id].cpu()
+ scale_coords(imgs[image_id].shape[1:], det[:, :4], img_shape).round()
+ scale_coords_landmarks(imgs[image_id].shape[1:], det[:, 5:15], img_shape).round()
+
+ for j in range(det.size()[0]):
+ box = (det[j, :4].view(1, 4) / gn).view(-1).tolist()
+ box = list(
+ map(int, [box[0] * image_width, box[1] * image_height, box[2] * image_width, box[3] * image_height])
+ )
+ if box[3] - box[1] < self.min_face:
+ continue
+ lm = (det[j, 5:15].view(1, 10) / gn_lks).view(-1).tolist()
+ lm = list(map(int, [i * image_width if j % 2 == 0 else i * image_height for j, i in enumerate(lm)]))
+ lm = [lm[i : i + 2] for i in range(0, len(lm), 2)]
+ bboxes[image_id].append(box)
+ landmarks[image_id].append(lm)
+ return bboxes, landmarks
+
+ def detect_faces(self, imgs, conf_thres=0.7, iou_thres=0.5):
+ """
+ Get bbox coordinates and keypoints of faces on original image.
+ Params:
+ imgs: image or list of images to detect faces on with BGR order (convert to RGB order for inference)
+ conf_thres: confidence threshold for each prediction
+ iou_thres: threshold for NMS (filter of intersecting bboxes)
+ Returns:
+ bboxes: list of arrays with 4 coordinates of bounding boxes with format x1,y1,x2,y2.
+ points: list of arrays with coordinates of 5 facial keypoints (eyes, nose, lips corners).
+ """
+ # Pass input images through face detector
+ images = imgs if isinstance(imgs, list) else [imgs]
+ images = [cv2.cvtColor(img, cv2.COLOR_BGR2RGB) for img in images]
+ origimgs = copy.deepcopy(images)
+
+ images = self._preprocess(images)
+
+ if IS_HIGH_VERSION:
+ with torch.inference_mode(): # for pytorch>=1.9
+ pred = self.detector(images)[0]
+ else:
+ with torch.no_grad(): # for pytorch<1.9
+ pred = self.detector(images)[0]
+
+ bboxes, points = self._postprocess(images, origimgs, pred, conf_thres, iou_thres)
+
+ # return bboxes, points
+ if not isListempty(points):
+ bboxes = np.array(bboxes).reshape(-1,4)
+ points = np.array(points).reshape(-1,10)
+ padding = bboxes[:,0].reshape(-1,1)
+ return np.concatenate((bboxes, padding, points), axis=1)
+ else:
+ return None
+
+ def __call__(self, *args):
+ return self.predict(*args)
diff --git a/r_facelib/detection/yolov5face/models/__init__.py b/r_facelib/detection/yolov5face/models/__init__.py new file mode 100644 index 0000000..e69de29 --- /dev/null +++ b/r_facelib/detection/yolov5face/models/__init__.py diff --git a/r_facelib/detection/yolov5face/models/common.py b/r_facelib/detection/yolov5face/models/common.py new file mode 100644 index 0000000..96894d5 --- /dev/null +++ b/r_facelib/detection/yolov5face/models/common.py @@ -0,0 +1,299 @@ +# This file contains modules common to various models
+
+import math
+
+import numpy as np
+import torch
+from torch import nn
+
+from r_facelib.detection.yolov5face.utils.datasets import letterbox
+from r_facelib.detection.yolov5face.utils.general import (
+ make_divisible,
+ non_max_suppression,
+ scale_coords,
+ xyxy2xywh,
+)
+
+
+def autopad(k, p=None): # kernel, padding
+ # Pad to 'same'
+ if p is None:
+ p = k // 2 if isinstance(k, int) else [x // 2 for x in k] # auto-pad
+ return p
+
+
+def channel_shuffle(x, groups):
+ batchsize, num_channels, height, width = x.data.size()
+ channels_per_group = torch.div(num_channels, groups, rounding_mode="trunc")
+
+ # reshape
+ x = x.view(batchsize, groups, channels_per_group, height, width)
+ x = torch.transpose(x, 1, 2).contiguous()
+
+ # flatten
+ return x.view(batchsize, -1, height, width)
+
+
+def DWConv(c1, c2, k=1, s=1, act=True):
+ # Depthwise convolution
+ return Conv(c1, c2, k, s, g=math.gcd(c1, c2), act=act)
+
+
+class Conv(nn.Module):
+ # Standard convolution
+ def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True): # ch_in, ch_out, kernel, stride, padding, groups
+ super().__init__()
+ self.conv = nn.Conv2d(c1, c2, k, s, autopad(k, p), groups=g, bias=False)
+ self.bn = nn.BatchNorm2d(c2)
+ self.act = nn.SiLU() if act is True else (act if isinstance(act, nn.Module) else nn.Identity())
+
+ def forward(self, x):
+ return self.act(self.bn(self.conv(x)))
+
+ def fuseforward(self, x):
+ return self.act(self.conv(x))
+
+
+class StemBlock(nn.Module):
+ def __init__(self, c1, c2, k=3, s=2, p=None, g=1, act=True):
+ super().__init__()
+ self.stem_1 = Conv(c1, c2, k, s, p, g, act)
+ self.stem_2a = Conv(c2, c2 // 2, 1, 1, 0)
+ self.stem_2b = Conv(c2 // 2, c2, 3, 2, 1)
+ self.stem_2p = nn.MaxPool2d(kernel_size=2, stride=2, ceil_mode=True)
+ self.stem_3 = Conv(c2 * 2, c2, 1, 1, 0)
+
+ def forward(self, x):
+ stem_1_out = self.stem_1(x)
+ stem_2a_out = self.stem_2a(stem_1_out)
+ stem_2b_out = self.stem_2b(stem_2a_out)
+ stem_2p_out = self.stem_2p(stem_1_out)
+ return self.stem_3(torch.cat((stem_2b_out, stem_2p_out), 1))
+
+
+class Bottleneck(nn.Module):
+ # Standard bottleneck
+ def __init__(self, c1, c2, shortcut=True, g=1, e=0.5): # ch_in, ch_out, shortcut, groups, expansion
+ super().__init__()
+ c_ = int(c2 * e) # hidden channels
+ self.cv1 = Conv(c1, c_, 1, 1)
+ self.cv2 = Conv(c_, c2, 3, 1, g=g)
+ self.add = shortcut and c1 == c2
+
+ def forward(self, x):
+ return x + self.cv2(self.cv1(x)) if self.add else self.cv2(self.cv1(x))
+
+
+class BottleneckCSP(nn.Module):
+ # CSP Bottleneck https://github.com/WongKinYiu/CrossStagePartialNetworks
+ def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5): # ch_in, ch_out, number, shortcut, groups, expansion
+ super().__init__()
+ c_ = int(c2 * e) # hidden channels
+ self.cv1 = Conv(c1, c_, 1, 1)
+ self.cv2 = nn.Conv2d(c1, c_, 1, 1, bias=False)
+ self.cv3 = nn.Conv2d(c_, c_, 1, 1, bias=False)
+ self.cv4 = Conv(2 * c_, c2, 1, 1)
+ self.bn = nn.BatchNorm2d(2 * c_) # applied to cat(cv2, cv3)
+ self.act = nn.LeakyReLU(0.1, inplace=True)
+ self.m = nn.Sequential(*(Bottleneck(c_, c_, shortcut, g, e=1.0) for _ in range(n)))
+
+ def forward(self, x):
+ y1 = self.cv3(self.m(self.cv1(x)))
+ y2 = self.cv2(x)
+ return self.cv4(self.act(self.bn(torch.cat((y1, y2), dim=1))))
+
+
+class C3(nn.Module):
+ # CSP Bottleneck with 3 convolutions
+ def __init__(self, c1, c2, n=1, shortcut=True, g=1, e=0.5): # ch_in, ch_out, number, shortcut, groups, expansion
+ super().__init__()
+ c_ = int(c2 * e) # hidden channels
+ self.cv1 = Conv(c1, c_, 1, 1)
+ self.cv2 = Conv(c1, c_, 1, 1)
+ self.cv3 = Conv(2 * c_, c2, 1) # act=FReLU(c2)
+ self.m = nn.Sequential(*(Bottleneck(c_, c_, shortcut, g, e=1.0) for _ in range(n)))
+
+ def forward(self, x):
+ return self.cv3(torch.cat((self.m(self.cv1(x)), self.cv2(x)), dim=1))
+
+
+class ShuffleV2Block(nn.Module):
+ def __init__(self, inp, oup, stride):
+ super().__init__()
+
+ if not 1 <= stride <= 3:
+ raise ValueError("illegal stride value")
+ self.stride = stride
+
+ branch_features = oup // 2
+
+ if self.stride > 1:
+ self.branch1 = nn.Sequential(
+ self.depthwise_conv(inp, inp, kernel_size=3, stride=self.stride, padding=1),
+ nn.BatchNorm2d(inp),
+ nn.Conv2d(inp, branch_features, kernel_size=1, stride=1, padding=0, bias=False),
+ nn.BatchNorm2d(branch_features),
+ nn.SiLU(),
+ )
+ else:
+ self.branch1 = nn.Sequential()
+
+ self.branch2 = nn.Sequential(
+ nn.Conv2d(
+ inp if (self.stride > 1) else branch_features,
+ branch_features,
+ kernel_size=1,
+ stride=1,
+ padding=0,
+ bias=False,
+ ),
+ nn.BatchNorm2d(branch_features),
+ nn.SiLU(),
+ self.depthwise_conv(branch_features, branch_features, kernel_size=3, stride=self.stride, padding=1),
+ nn.BatchNorm2d(branch_features),
+ nn.Conv2d(branch_features, branch_features, kernel_size=1, stride=1, padding=0, bias=False),
+ nn.BatchNorm2d(branch_features),
+ nn.SiLU(),
+ )
+
+ @staticmethod
+ def depthwise_conv(i, o, kernel_size, stride=1, padding=0, bias=False):
+ return nn.Conv2d(i, o, kernel_size, stride, padding, bias=bias, groups=i)
+
+ def forward(self, x):
+ if self.stride == 1:
+ x1, x2 = x.chunk(2, dim=1)
+ out = torch.cat((x1, self.branch2(x2)), dim=1)
+ else:
+ out = torch.cat((self.branch1(x), self.branch2(x)), dim=1)
+ out = channel_shuffle(out, 2)
+ return out
+
+
+class SPP(nn.Module):
+ # Spatial pyramid pooling layer used in YOLOv3-SPP
+ def __init__(self, c1, c2, k=(5, 9, 13)):
+ super().__init__()
+ c_ = c1 // 2 # hidden channels
+ self.cv1 = Conv(c1, c_, 1, 1)
+ self.cv2 = Conv(c_ * (len(k) + 1), c2, 1, 1)
+ self.m = nn.ModuleList([nn.MaxPool2d(kernel_size=x, stride=1, padding=x // 2) for x in k])
+
+ def forward(self, x):
+ x = self.cv1(x)
+ return self.cv2(torch.cat([x] + [m(x) for m in self.m], 1))
+
+
+class Focus(nn.Module):
+ # Focus wh information into c-space
+ def __init__(self, c1, c2, k=1, s=1, p=None, g=1, act=True): # ch_in, ch_out, kernel, stride, padding, groups
+ super().__init__()
+ self.conv = Conv(c1 * 4, c2, k, s, p, g, act)
+
+ def forward(self, x): # x(b,c,w,h) -> y(b,4c,w/2,h/2)
+ return self.conv(torch.cat([x[..., ::2, ::2], x[..., 1::2, ::2], x[..., ::2, 1::2], x[..., 1::2, 1::2]], 1))
+
+
+class Concat(nn.Module):
+ # Concatenate a list of tensors along dimension
+ def __init__(self, dimension=1):
+ super().__init__()
+ self.d = dimension
+
+ def forward(self, x):
+ return torch.cat(x, self.d)
+
+
+class NMS(nn.Module):
+ # Non-Maximum Suppression (NMS) module
+ conf = 0.25 # confidence threshold
+ iou = 0.45 # IoU threshold
+ classes = None # (optional list) filter by class
+
+ def forward(self, x):
+ return non_max_suppression(x[0], conf_thres=self.conf, iou_thres=self.iou, classes=self.classes)
+
+
+class AutoShape(nn.Module):
+ # input-robust model wrapper for passing cv2/np/PIL/torch inputs. Includes preprocessing, inference and NMS
+ img_size = 640 # inference size (pixels)
+ conf = 0.25 # NMS confidence threshold
+ iou = 0.45 # NMS IoU threshold
+ classes = None # (optional list) filter by class
+
+ def __init__(self, model):
+ super().__init__()
+ self.model = model.eval()
+
+ def autoshape(self):
+ print("autoShape already enabled, skipping... ") # model already converted to model.autoshape()
+ return self
+
+ def forward(self, imgs, size=640, augment=False, profile=False):
+ # Inference from various sources. For height=720, width=1280, RGB images example inputs are:
+ # OpenCV: = cv2.imread('image.jpg')[:,:,::-1] # HWC BGR to RGB x(720,1280,3)
+ # PIL: = Image.open('image.jpg') # HWC x(720,1280,3)
+ # numpy: = np.zeros((720,1280,3)) # HWC
+ # torch: = torch.zeros(16,3,720,1280) # BCHW
+ # multiple: = [Image.open('image1.jpg'), Image.open('image2.jpg'), ...] # list of images
+
+ p = next(self.model.parameters()) # for device and type
+ if isinstance(imgs, torch.Tensor): # torch
+ return self.model(imgs.to(p.device).type_as(p), augment, profile) # inference
+
+ # Pre-process
+ n, imgs = (len(imgs), imgs) if isinstance(imgs, list) else (1, [imgs]) # number of images, list of images
+ shape0, shape1 = [], [] # image and inference shapes
+ for i, im in enumerate(imgs):
+ im = np.array(im) # to numpy
+ if im.shape[0] < 5: # image in CHW
+ im = im.transpose((1, 2, 0)) # reverse dataloader .transpose(2, 0, 1)
+ im = im[:, :, :3] if im.ndim == 3 else np.tile(im[:, :, None], 3) # enforce 3ch input
+ s = im.shape[:2] # HWC
+ shape0.append(s) # image shape
+ g = size / max(s) # gain
+ shape1.append([y * g for y in s])
+ imgs[i] = im # update
+ shape1 = [make_divisible(x, int(self.stride.max())) for x in np.stack(shape1, 0).max(0)] # inference shape
+ x = [letterbox(im, new_shape=shape1, auto=False)[0] for im in imgs] # pad
+ x = np.stack(x, 0) if n > 1 else x[0][None] # stack
+ x = np.ascontiguousarray(x.transpose((0, 3, 1, 2))) # BHWC to BCHW
+ x = torch.from_numpy(x).to(p.device).type_as(p) / 255.0 # uint8 to fp16/32
+
+ # Inference
+ with torch.no_grad():
+ y = self.model(x, augment, profile)[0] # forward
+ y = non_max_suppression(y, conf_thres=self.conf, iou_thres=self.iou, classes=self.classes) # NMS
+
+ # Post-process
+ for i in range(n):
+ scale_coords(shape1, y[i][:, :4], shape0[i])
+
+ return Detections(imgs, y, self.names)
+
+
+class Detections:
+ # detections class for YOLOv5 inference results
+ def __init__(self, imgs, pred, names=None):
+ super().__init__()
+ d = pred[0].device # device
+ gn = [torch.tensor([*(im.shape[i] for i in [1, 0, 1, 0]), 1.0, 1.0], device=d) for im in imgs] # normalizations
+ self.imgs = imgs # list of images as numpy arrays
+ self.pred = pred # list of tensors pred[0] = (xyxy, conf, cls)
+ self.names = names # class names
+ self.xyxy = pred # xyxy pixels
+ self.xywh = [xyxy2xywh(x) for x in pred] # xywh pixels
+ self.xyxyn = [x / g for x, g in zip(self.xyxy, gn)] # xyxy normalized
+ self.xywhn = [x / g for x, g in zip(self.xywh, gn)] # xywh normalized
+ self.n = len(self.pred)
+
+ def __len__(self):
+ return self.n
+
+ def tolist(self):
+ # return a list of Detections objects, i.e. 'for result in results.tolist():'
+ x = [Detections([self.imgs[i]], [self.pred[i]], self.names) for i in range(self.n)]
+ for d in x:
+ for k in ["imgs", "pred", "xyxy", "xyxyn", "xywh", "xywhn"]:
+ setattr(d, k, getattr(d, k)[0]) # pop out of list
+ return x
diff --git a/r_facelib/detection/yolov5face/models/experimental.py b/r_facelib/detection/yolov5face/models/experimental.py new file mode 100644 index 0000000..bdf7aea --- /dev/null +++ b/r_facelib/detection/yolov5face/models/experimental.py @@ -0,0 +1,45 @@ +# # This file contains experimental modules
+
+import numpy as np
+import torch
+from torch import nn
+
+from r_facelib.detection.yolov5face.models.common import Conv
+
+
+class CrossConv(nn.Module):
+ # Cross Convolution Downsample
+ def __init__(self, c1, c2, k=3, s=1, g=1, e=1.0, shortcut=False):
+ # ch_in, ch_out, kernel, stride, groups, expansion, shortcut
+ super().__init__()
+ c_ = int(c2 * e) # hidden channels
+ self.cv1 = Conv(c1, c_, (1, k), (1, s))
+ self.cv2 = Conv(c_, c2, (k, 1), (s, 1), g=g)
+ self.add = shortcut and c1 == c2
+
+ def forward(self, x):
+ return x + self.cv2(self.cv1(x)) if self.add else self.cv2(self.cv1(x))
+
+
+class MixConv2d(nn.Module):
+ # Mixed Depthwise Conv https://arxiv.org/abs/1907.09595
+ def __init__(self, c1, c2, k=(1, 3), s=1, equal_ch=True):
+ super().__init__()
+ groups = len(k)
+ if equal_ch: # equal c_ per group
+ i = torch.linspace(0, groups - 1e-6, c2).floor() # c2 indices
+ c_ = [(i == g).sum() for g in range(groups)] # intermediate channels
+ else: # equal weight.numel() per group
+ b = [c2] + [0] * groups
+ a = np.eye(groups + 1, groups, k=-1)
+ a -= np.roll(a, 1, axis=1)
+ a *= np.array(k) ** 2
+ a[0] = 1
+ c_ = np.linalg.lstsq(a, b, rcond=None)[0].round() # solve for equal weight indices, ax = b
+
+ self.m = nn.ModuleList([nn.Conv2d(c1, int(c_[g]), k[g], s, k[g] // 2, bias=False) for g in range(groups)])
+ self.bn = nn.BatchNorm2d(c2)
+ self.act = nn.LeakyReLU(0.1, inplace=True)
+
+ def forward(self, x):
+ return x + self.act(self.bn(torch.cat([m(x) for m in self.m], 1)))
diff --git a/r_facelib/detection/yolov5face/models/yolo.py b/r_facelib/detection/yolov5face/models/yolo.py new file mode 100644 index 0000000..02479dc --- /dev/null +++ b/r_facelib/detection/yolov5face/models/yolo.py @@ -0,0 +1,235 @@ +import math
+from copy import deepcopy
+from pathlib import Path
+
+import torch
+import yaml # for torch hub
+from torch import nn
+
+from r_facelib.detection.yolov5face.models.common import (
+ C3,
+ NMS,
+ SPP,
+ AutoShape,
+ Bottleneck,
+ BottleneckCSP,
+ Concat,
+ Conv,
+ DWConv,
+ Focus,
+ ShuffleV2Block,
+ StemBlock,
+)
+from r_facelib.detection.yolov5face.models.experimental import CrossConv, MixConv2d
+from r_facelib.detection.yolov5face.utils.autoanchor import check_anchor_order
+from r_facelib.detection.yolov5face.utils.general import make_divisible
+from r_facelib.detection.yolov5face.utils.torch_utils import copy_attr, fuse_conv_and_bn
+
+
+class Detect(nn.Module):
+ stride = None # strides computed during build
+ export = False # onnx export
+
+ def __init__(self, nc=80, anchors=(), ch=()): # detection layer
+ super().__init__()
+ self.nc = nc # number of classes
+ self.no = nc + 5 + 10 # number of outputs per anchor
+
+ self.nl = len(anchors) # number of detection layers
+ self.na = len(anchors[0]) // 2 # number of anchors
+ self.grid = [torch.zeros(1)] * self.nl # init grid
+ a = torch.tensor(anchors).float().view(self.nl, -1, 2)
+ self.register_buffer("anchors", a) # shape(nl,na,2)
+ self.register_buffer("anchor_grid", a.clone().view(self.nl, 1, -1, 1, 1, 2)) # shape(nl,1,na,1,1,2)
+ self.m = nn.ModuleList(nn.Conv2d(x, self.no * self.na, 1) for x in ch) # output conv
+
+ def forward(self, x):
+ z = [] # inference output
+ if self.export:
+ for i in range(self.nl):
+ x[i] = self.m[i](x[i])
+ return x
+ for i in range(self.nl):
+ x[i] = self.m[i](x[i]) # conv
+ bs, _, 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: # inference
+ if self.grid[i].shape[2:4] != x[i].shape[2:4]:
+ self.grid[i] = self._make_grid(nx, ny).to(x[i].device)
+
+ y = torch.full_like(x[i], 0)
+ y[..., [0, 1, 2, 3, 4, 15]] = x[i][..., [0, 1, 2, 3, 4, 15]].sigmoid()
+ y[..., 5:15] = x[i][..., 5:15]
+
+ y[..., 0:2] = (y[..., 0:2] * 2.0 - 0.5 + self.grid[i].to(x[i].device)) * self.stride[i] # xy
+ y[..., 2:4] = (y[..., 2:4] * 2) ** 2 * self.anchor_grid[i] # wh
+
+ y[..., 5:7] = (
+ y[..., 5:7] * self.anchor_grid[i] + self.grid[i].to(x[i].device) * self.stride[i]
+ ) # landmark x1 y1
+ y[..., 7:9] = (
+ y[..., 7:9] * self.anchor_grid[i] + self.grid[i].to(x[i].device) * self.stride[i]
+ ) # landmark x2 y2
+ y[..., 9:11] = (
+ y[..., 9:11] * self.anchor_grid[i] + self.grid[i].to(x[i].device) * self.stride[i]
+ ) # landmark x3 y3
+ y[..., 11:13] = (
+ y[..., 11:13] * self.anchor_grid[i] + self.grid[i].to(x[i].device) * self.stride[i]
+ ) # landmark x4 y4
+ y[..., 13:15] = (
+ y[..., 13:15] * self.anchor_grid[i] + self.grid[i].to(x[i].device) * self.stride[i]
+ ) # landmark x5 y5
+
+ z.append(y.view(bs, -1, self.no))
+
+ return x if self.training else (torch.cat(z, 1), x)
+
+ @staticmethod
+ def _make_grid(nx=20, ny=20):
+ # yv, xv = torch.meshgrid([torch.arange(ny), torch.arange(nx)], indexing="ij") # for pytorch>=1.10
+ yv, xv = torch.meshgrid([torch.arange(ny), torch.arange(nx)])
+ return torch.stack((xv, yv), 2).view((1, 1, ny, nx, 2)).float()
+
+
+class Model(nn.Module):
+ def __init__(self, cfg="yolov5s.yaml", ch=3, nc=None): # model, input channels, number of classes
+ super().__init__()
+ self.yaml_file = Path(cfg).name
+ with Path(cfg).open(encoding="utf8") as f:
+ self.yaml = yaml.safe_load(f) # model dict
+
+ # Define model
+ ch = self.yaml["ch"] = self.yaml.get("ch", ch) # input channels
+ if nc and nc != self.yaml["nc"]:
+ self.yaml["nc"] = nc # override yaml value
+
+ self.model, self.save = parse_model(deepcopy(self.yaml), ch=[ch]) # model, savelist
+ self.names = [str(i) for i in range(self.yaml["nc"])] # default names
+
+ # Build strides, anchors
+ m = self.model[-1] # Detect()
+ if isinstance(m, Detect):
+ s = 128 # 2x min stride
+ m.stride = torch.tensor([s / x.shape[-2] for x in self.forward(torch.zeros(1, ch, s, s))]) # forward
+ m.anchors /= m.stride.view(-1, 1, 1)
+ check_anchor_order(m)
+ self.stride = m.stride
+ self._initialize_biases() # only run once
+
+ def forward(self, x):
+ return self.forward_once(x) # single-scale inference, train
+
+ def forward_once(self, x):
+ y = [] # outputs
+ for m in self.model:
+ if m.f != -1: # if not from previous layer
+ x = y[m.f] if isinstance(m.f, int) else [x if j == -1 else y[j] for j in m.f] # from earlier layers
+
+ x = m(x) # run
+ y.append(x if m.i in self.save else None) # save output
+
+ return x
+
+ def _initialize_biases(self, cf=None): # initialize biases into Detect(), cf is class frequency
+ # https://arxiv.org/abs/1708.02002 section 3.3
+ m = self.model[-1] # Detect() module
+ for mi, s in zip(m.m, m.stride): # from
+ b = mi.bias.view(m.na, -1) # conv.bias(255) to (3,85)
+ b.data[:, 4] += math.log(8 / (640 / s) ** 2) # obj (8 objects per 640 image)
+ b.data[:, 5:] += math.log(0.6 / (m.nc - 0.99)) if cf is None else torch.log(cf / cf.sum()) # cls
+ mi.bias = torch.nn.Parameter(b.view(-1), requires_grad=True)
+
+ def _print_biases(self):
+ m = self.model[-1] # Detect() module
+ for mi in m.m: # from
+ b = mi.bias.detach().view(m.na, -1).T # conv.bias(255) to (3,85)
+ print(("%6g Conv2d.bias:" + "%10.3g" * 6) % (mi.weight.shape[1], *b[:5].mean(1).tolist(), b[5:].mean()))
+
+ def fuse(self): # fuse model Conv2d() + BatchNorm2d() layers
+ print("Fusing layers... ")
+ for m in self.model.modules():
+ if isinstance(m, Conv) and hasattr(m, "bn"):
+ m.conv = fuse_conv_and_bn(m.conv, m.bn) # update conv
+ delattr(m, "bn") # remove batchnorm
+ m.forward = m.fuseforward # update forward
+ elif type(m) is nn.Upsample:
+ m.recompute_scale_factor = None # torch 1.11.0 compatibility
+ return self
+
+ def nms(self, mode=True): # add or remove NMS module
+ present = isinstance(self.model[-1], NMS) # last layer is NMS
+ if mode and not present:
+ print("Adding NMS... ")
+ m = NMS() # module
+ m.f = -1 # from
+ m.i = self.model[-1].i + 1 # index
+ self.model.add_module(name=str(m.i), module=m) # add
+ self.eval()
+ elif not mode and present:
+ print("Removing NMS... ")
+ self.model = self.model[:-1] # remove
+ return self
+
+ def autoshape(self): # add autoShape module
+ print("Adding autoShape... ")
+ m = AutoShape(self) # wrap model
+ copy_attr(m, self, include=("yaml", "nc", "hyp", "names", "stride"), exclude=()) # copy attributes
+ return m
+
+
+def parse_model(d, ch): # model_dict, input_channels(3)
+ anchors, nc, gd, gw = d["anchors"], d["nc"], d["depth_multiple"], d["width_multiple"]
+ na = (len(anchors[0]) // 2) if isinstance(anchors, list) else anchors # number of anchors
+ no = na * (nc + 5) # number of outputs = anchors * (classes + 5)
+
+ layers, save, c2 = [], [], ch[-1] # layers, savelist, ch out
+ for i, (f, n, m, args) in enumerate(d["backbone"] + d["head"]): # from, number, module, args
+ m = eval(m) if isinstance(m, str) else m # eval strings
+ for j, a in enumerate(args):
+ try:
+ args[j] = eval(a) if isinstance(a, str) else a # eval strings
+ except:
+ pass
+
+ n = max(round(n * gd), 1) if n > 1 else n # depth gain
+ if m in [
+ Conv,
+ Bottleneck,
+ SPP,
+ DWConv,
+ MixConv2d,
+ Focus,
+ CrossConv,
+ BottleneckCSP,
+ C3,
+ ShuffleV2Block,
+ StemBlock,
+ ]:
+ c1, c2 = ch[f], args[0]
+
+ c2 = make_divisible(c2 * gw, 8) if c2 != no else c2
+
+ args = [c1, c2, *args[1:]]
+ if m in [BottleneckCSP, C3]:
+ args.insert(2, n)
+ n = 1
+ elif m is nn.BatchNorm2d:
+ args = [ch[f]]
+ elif m is Concat:
+ c2 = sum(ch[-1 if x == -1 else x + 1] for x in f)
+ elif m is Detect:
+ args.append([ch[x + 1] for x in f])
+ if isinstance(args[1], int): # number of anchors
+ args[1] = [list(range(args[1] * 2))] * len(f)
+ else:
+ c2 = ch[f]
+
+ m_ = nn.Sequential(*(m(*args) for _ in range(n))) if n > 1 else m(*args) # module
+ t = str(m)[8:-2].replace("__main__.", "") # module type
+ np = sum(x.numel() for x in m_.parameters()) # number params
+ m_.i, m_.f, m_.type, m_.np = i, f, t, np # attach index, 'from' index, type, number params
+ save.extend(x % i for x in ([f] if isinstance(f, int) else f) if x != -1) # append to savelist
+ layers.append(m_)
+ ch.append(c2)
+ return nn.Sequential(*layers), sorted(save)
diff --git a/r_facelib/detection/yolov5face/models/yolov5l.yaml b/r_facelib/detection/yolov5face/models/yolov5l.yaml new file mode 100644 index 0000000..98a9e2c --- /dev/null +++ b/r_facelib/detection/yolov5face/models/yolov5l.yaml @@ -0,0 +1,47 @@ +# parameters
+nc: 1 # number of classes
+depth_multiple: 1.0 # model depth multiple
+width_multiple: 1.0 # layer channel multiple
+
+# anchors
+anchors:
+ - [4,5, 8,10, 13,16] # P3/8
+ - [23,29, 43,55, 73,105] # P4/16
+ - [146,217, 231,300, 335,433] # P5/32
+
+# YOLOv5 backbone
+backbone:
+ # [from, number, module, args]
+ [[-1, 1, StemBlock, [64, 3, 2]], # 0-P1/2
+ [-1, 3, C3, [128]],
+ [-1, 1, Conv, [256, 3, 2]], # 2-P3/8
+ [-1, 9, C3, [256]],
+ [-1, 1, Conv, [512, 3, 2]], # 4-P4/16
+ [-1, 9, C3, [512]],
+ [-1, 1, Conv, [1024, 3, 2]], # 6-P5/32
+ [-1, 1, SPP, [1024, [3,5,7]]],
+ [-1, 3, C3, [1024, False]], # 8
+ ]
+
+# YOLOv5 head
+head:
+ [[-1, 1, Conv, [512, 1, 1]],
+ [-1, 1, nn.Upsample, [None, 2, 'nearest']],
+ [[-1, 5], 1, Concat, [1]], # cat backbone P4
+ [-1, 3, C3, [512, False]], # 12
+
+ [-1, 1, Conv, [256, 1, 1]],
+ [-1, 1, nn.Upsample, [None, 2, 'nearest']],
+ [[-1, 3], 1, Concat, [1]], # cat backbone P3
+ [-1, 3, C3, [256, False]], # 16 (P3/8-small)
+
+ [-1, 1, Conv, [256, 3, 2]],
+ [[-1, 13], 1, Concat, [1]], # cat head P4
+ [-1, 3, C3, [512, False]], # 19 (P4/16-medium)
+
+ [-1, 1, Conv, [512, 3, 2]],
+ [[-1, 9], 1, Concat, [1]], # cat head P5
+ [-1, 3, C3, [1024, False]], # 22 (P5/32-large)
+
+ [[16, 19, 22], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5)
+ ]
\ No newline at end of file diff --git a/r_facelib/detection/yolov5face/models/yolov5n.yaml b/r_facelib/detection/yolov5face/models/yolov5n.yaml new file mode 100644 index 0000000..0a03fb0 --- /dev/null +++ b/r_facelib/detection/yolov5face/models/yolov5n.yaml @@ -0,0 +1,45 @@ +# parameters
+nc: 1 # number of classes
+depth_multiple: 1.0 # model depth multiple
+width_multiple: 1.0 # layer channel multiple
+
+# anchors
+anchors:
+ - [4,5, 8,10, 13,16] # P3/8
+ - [23,29, 43,55, 73,105] # P4/16
+ - [146,217, 231,300, 335,433] # P5/32
+
+# YOLOv5 backbone
+backbone:
+ # [from, number, module, args]
+ [[-1, 1, StemBlock, [32, 3, 2]], # 0-P2/4
+ [-1, 1, ShuffleV2Block, [128, 2]], # 1-P3/8
+ [-1, 3, ShuffleV2Block, [128, 1]], # 2
+ [-1, 1, ShuffleV2Block, [256, 2]], # 3-P4/16
+ [-1, 7, ShuffleV2Block, [256, 1]], # 4
+ [-1, 1, ShuffleV2Block, [512, 2]], # 5-P5/32
+ [-1, 3, ShuffleV2Block, [512, 1]], # 6
+ ]
+
+# YOLOv5 head
+head:
+ [[-1, 1, Conv, [128, 1, 1]],
+ [-1, 1, nn.Upsample, [None, 2, 'nearest']],
+ [[-1, 4], 1, Concat, [1]], # cat backbone P4
+ [-1, 1, C3, [128, False]], # 10
+
+ [-1, 1, Conv, [128, 1, 1]],
+ [-1, 1, nn.Upsample, [None, 2, 'nearest']],
+ [[-1, 2], 1, Concat, [1]], # cat backbone P3
+ [-1, 1, C3, [128, False]], # 14 (P3/8-small)
+
+ [-1, 1, Conv, [128, 3, 2]],
+ [[-1, 11], 1, Concat, [1]], # cat head P4
+ [-1, 1, C3, [128, False]], # 17 (P4/16-medium)
+
+ [-1, 1, Conv, [128, 3, 2]],
+ [[-1, 7], 1, Concat, [1]], # cat head P5
+ [-1, 1, C3, [128, False]], # 20 (P5/32-large)
+
+ [[14, 17, 20], 1, Detect, [nc, anchors]], # Detect(P3, P4, P5)
+ ]
diff --git a/r_facelib/detection/yolov5face/utils/__init__.py b/r_facelib/detection/yolov5face/utils/__init__.py new file mode 100644 index 0000000..e69de29 --- /dev/null +++ b/r_facelib/detection/yolov5face/utils/__init__.py diff --git a/r_facelib/detection/yolov5face/utils/autoanchor.py b/r_facelib/detection/yolov5face/utils/autoanchor.py new file mode 100644 index 0000000..cb0de89 --- /dev/null +++ b/r_facelib/detection/yolov5face/utils/autoanchor.py @@ -0,0 +1,12 @@ +# Auto-anchor utils
+
+
+def check_anchor_order(m):
+ # Check anchor order against stride order for YOLOv5 Detect() module m, and correct if necessary
+ a = m.anchor_grid.prod(-1).view(-1) # anchor area
+ da = a[-1] - a[0] # delta a
+ ds = m.stride[-1] - m.stride[0] # delta s
+ if da.sign() != ds.sign(): # same order
+ print("Reversing anchor order")
+ m.anchors[:] = m.anchors.flip(0)
+ m.anchor_grid[:] = m.anchor_grid.flip(0)
diff --git a/r_facelib/detection/yolov5face/utils/datasets.py b/r_facelib/detection/yolov5face/utils/datasets.py new file mode 100644 index 0000000..a72609b --- /dev/null +++ b/r_facelib/detection/yolov5face/utils/datasets.py @@ -0,0 +1,35 @@ +import cv2
+import numpy as np
+
+
+def letterbox(img, new_shape=(640, 640), color=(114, 114, 114), auto=True, scale_fill=False, scaleup=True):
+ # Resize image to a 32-pixel-multiple rectangle https://github.com/ultralytics/yolov3/issues/232
+ shape = img.shape[:2] # current shape [height, width]
+ if isinstance(new_shape, int):
+ new_shape = (new_shape, new_shape)
+
+ # Scale ratio (new / old)
+ r = min(new_shape[0] / shape[0], new_shape[1] / shape[1])
+ if not scaleup: # only scale down, do not scale up (for better test mAP)
+ r = min(r, 1.0)
+
+ # Compute padding
+ ratio = r, r # width, height ratios
+ new_unpad = int(round(shape[1] * r)), int(round(shape[0] * r))
+ dw, dh = new_shape[1] - new_unpad[0], new_shape[0] - new_unpad[1] # wh padding
+ if auto: # minimum rectangle
+ dw, dh = np.mod(dw, 64), np.mod(dh, 64) # wh padding
+ elif scale_fill: # stretch
+ dw, dh = 0.0, 0.0
+ new_unpad = (new_shape[1], new_shape[0])
+ ratio = new_shape[1] / shape[1], new_shape[0] / shape[0] # width, height ratios
+
+ dw /= 2 # divide padding into 2 sides
+ dh /= 2
+
+ if shape[::-1] != new_unpad: # resize
+ img = cv2.resize(img, new_unpad, interpolation=cv2.INTER_LINEAR)
+ top, bottom = int(round(dh - 0.1)), int(round(dh + 0.1))
+ left, right = int(round(dw - 0.1)), int(round(dw + 0.1))
+ img = cv2.copyMakeBorder(img, top, bottom, left, right, cv2.BORDER_CONSTANT, value=color) # add border
+ return img, ratio, (dw, dh)
diff --git a/r_facelib/detection/yolov5face/utils/extract_ckpt.py b/r_facelib/detection/yolov5face/utils/extract_ckpt.py new file mode 100644 index 0000000..e6bde00 --- /dev/null +++ b/r_facelib/detection/yolov5face/utils/extract_ckpt.py @@ -0,0 +1,5 @@ +import torch
+import sys
+sys.path.insert(0,'./facelib/detection/yolov5face')
+model = torch.load('facelib/detection/yolov5face/yolov5n-face.pt', map_location='cpu')['model']
+torch.save(model.state_dict(),'../../models/facedetection')
\ No newline at end of file diff --git a/r_facelib/detection/yolov5face/utils/general.py b/r_facelib/detection/yolov5face/utils/general.py new file mode 100644 index 0000000..618d2f3 --- /dev/null +++ b/r_facelib/detection/yolov5face/utils/general.py @@ -0,0 +1,271 @@ +import math
+import time
+
+import numpy as np
+import torch
+import torchvision
+
+
+def check_img_size(img_size, s=32):
+ # Verify img_size is a multiple of stride s
+ new_size = make_divisible(img_size, int(s)) # ceil gs-multiple
+ # if new_size != img_size:
+ # print(f"WARNING: --img-size {img_size:g} must be multiple of max stride {s:g}, updating to {new_size:g}")
+ return new_size
+
+
+def make_divisible(x, divisor):
+ # Returns x evenly divisible by divisor
+ return math.ceil(x / divisor) * divisor
+
+
+def xyxy2xywh(x):
+ # Convert nx4 boxes from [x1, y1, x2, y2] to [x, y, w, h] where xy1=top-left, xy2=bottom-right
+ y = x.clone() if isinstance(x, torch.Tensor) else np.copy(x)
+ y[:, 0] = (x[:, 0] + x[:, 2]) / 2 # x center
+ y[:, 1] = (x[:, 1] + x[:, 3]) / 2 # y center
+ y[:, 2] = x[:, 2] - x[:, 0] # width
+ y[:, 3] = x[:, 3] - x[:, 1] # height
+ return y
+
+
+def xywh2xyxy(x):
+ # Convert nx4 boxes from [x, y, w, h] to [x1, y1, x2, y2] where xy1=top-left, xy2=bottom-right
+ y = x.clone() if isinstance(x, torch.Tensor) else np.copy(x)
+ y[:, 0] = x[:, 0] - x[:, 2] / 2 # top left x
+ y[:, 1] = x[:, 1] - x[:, 3] / 2 # top left y
+ y[:, 2] = x[:, 0] + x[:, 2] / 2 # bottom right x
+ y[:, 3] = x[:, 1] + x[:, 3] / 2 # bottom right y
+ return y
+
+
+def scale_coords(img1_shape, coords, img0_shape, ratio_pad=None):
+ # Rescale coords (xyxy) from img1_shape to img0_shape
+ if ratio_pad is None: # calculate from img0_shape
+ gain = min(img1_shape[0] / img0_shape[0], img1_shape[1] / img0_shape[1]) # gain = old / new
+ pad = (img1_shape[1] - img0_shape[1] * gain) / 2, (img1_shape[0] - img0_shape[0] * gain) / 2 # wh padding
+ else:
+ gain = ratio_pad[0][0]
+ pad = ratio_pad[1]
+
+ coords[:, [0, 2]] -= pad[0] # x padding
+ coords[:, [1, 3]] -= pad[1] # y padding
+ coords[:, :4] /= gain
+ clip_coords(coords, img0_shape)
+ return coords
+
+
+def clip_coords(boxes, img_shape):
+ # Clip bounding xyxy bounding boxes to image shape (height, width)
+ boxes[:, 0].clamp_(0, img_shape[1]) # x1
+ boxes[:, 1].clamp_(0, img_shape[0]) # y1
+ boxes[:, 2].clamp_(0, img_shape[1]) # x2
+ boxes[:, 3].clamp_(0, img_shape[0]) # y2
+
+
+def box_iou(box1, box2):
+ # https://github.com/pytorch/vision/blob/master/torchvision/ops/boxes.py
+ """
+ Return intersection-over-union (Jaccard index) of boxes.
+ Both sets of boxes are expected to be in (x1, y1, x2, y2) format.
+ Arguments:
+ box1 (Tensor[N, 4])
+ box2 (Tensor[M, 4])
+ Returns:
+ iou (Tensor[N, M]): the NxM matrix containing the pairwise
+ IoU values for every element in boxes1 and boxes2
+ """
+
+ def box_area(box):
+ return (box[2] - box[0]) * (box[3] - box[1])
+
+ area1 = box_area(box1.T)
+ area2 = box_area(box2.T)
+
+ inter = (torch.min(box1[:, None, 2:], box2[:, 2:]) - torch.max(box1[:, None, :2], box2[:, :2])).clamp(0).prod(2)
+ return inter / (area1[:, None] + area2 - inter)
+
+
+def non_max_suppression_face(prediction, conf_thres=0.25, iou_thres=0.45, classes=None, agnostic=False, labels=()):
+ """Performs Non-Maximum Suppression (NMS) on inference results
+ Returns:
+ detections with shape: nx6 (x1, y1, x2, y2, conf, cls)
+ """
+
+ nc = prediction.shape[2] - 15 # number of classes
+ xc = prediction[..., 4] > conf_thres # candidates
+
+ # Settings
+ # (pixels) maximum box width and height
+ max_wh = 4096
+ time_limit = 10.0 # seconds to quit after
+ redundant = True # require redundant detections
+ multi_label = nc > 1 # multiple labels per box (adds 0.5ms/img)
+ merge = False # use merge-NMS
+
+ t = time.time()
+ output = [torch.zeros((0, 16), device=prediction.device)] * prediction.shape[0]
+ for xi, x in enumerate(prediction): # image index, image inference
+ # Apply constraints
+ x = x[xc[xi]] # confidence
+
+ # Cat apriori labels if autolabelling
+ if labels and len(labels[xi]):
+ label = labels[xi]
+ v = torch.zeros((len(label), nc + 15), device=x.device)
+ v[:, :4] = label[:, 1:5] # box
+ v[:, 4] = 1.0 # conf
+ v[range(len(label)), label[:, 0].long() + 15] = 1.0 # cls
+ x = torch.cat((x, v), 0)
+
+ # If none remain process next image
+ if not x.shape[0]:
+ continue
+
+ # Compute conf
+ x[:, 15:] *= x[:, 4:5] # conf = obj_conf * cls_conf
+
+ # Box (center x, center y, width, height) to (x1, y1, x2, y2)
+ box = xywh2xyxy(x[:, :4])
+
+ # Detections matrix nx6 (xyxy, conf, landmarks, cls)
+ if multi_label:
+ i, j = (x[:, 15:] > conf_thres).nonzero(as_tuple=False).T
+ x = torch.cat((box[i], x[i, j + 15, None], x[:, 5:15], j[:, None].float()), 1)
+ else: # best class only
+ conf, j = x[:, 15:].max(1, keepdim=True)
+ x = torch.cat((box, conf, x[:, 5:15], j.float()), 1)[conf.view(-1) > conf_thres]
+
+ # Filter by class
+ if classes is not None:
+ x = x[(x[:, 5:6] == torch.tensor(classes, device=x.device)).any(1)]
+
+ # If none remain process next image
+ n = x.shape[0] # number of boxes
+ if not n:
+ continue
+
+ # Batched NMS
+ c = x[:, 15:16] * (0 if agnostic else max_wh) # classes
+ boxes, scores = x[:, :4] + c, x[:, 4] # boxes (offset by class), scores
+ i = torchvision.ops.nms(boxes, scores, iou_thres) # NMS
+
+ if merge and (1 < n < 3e3): # Merge NMS (boxes merged using weighted mean)
+ # update boxes as boxes(i,4) = weights(i,n) * boxes(n,4)
+ iou = box_iou(boxes[i], boxes) > iou_thres # iou matrix
+ weights = iou * scores[None] # box weights
+ x[i, :4] = torch.mm(weights, x[:, :4]).float() / weights.sum(1, keepdim=True) # merged boxes
+ if redundant:
+ i = i[iou.sum(1) > 1] # require redundancy
+
+ output[xi] = x[i]
+ if (time.time() - t) > time_limit:
+ break # time limit exceeded
+
+ return output
+
+
+def non_max_suppression(prediction, conf_thres=0.25, iou_thres=0.45, classes=None, agnostic=False, labels=()):
+ """Performs Non-Maximum Suppression (NMS) on inference results
+
+ Returns:
+ detections with shape: nx6 (x1, y1, x2, y2, conf, cls)
+ """
+
+ nc = prediction.shape[2] - 5 # number of classes
+ xc = prediction[..., 4] > conf_thres # candidates
+
+ # Settings
+ # (pixels) maximum box width and height
+ max_wh = 4096
+ time_limit = 10.0 # seconds to quit after
+ redundant = True # require redundant detections
+ multi_label = nc > 1 # multiple labels per box (adds 0.5ms/img)
+ merge = False # use merge-NMS
+
+ t = time.time()
+ output = [torch.zeros((0, 6), device=prediction.device)] * prediction.shape[0]
+ for xi, x in enumerate(prediction): # image index, image inference
+ x = x[xc[xi]] # confidence
+
+ # Cat apriori labels if autolabelling
+ if labels and len(labels[xi]):
+ label_id = labels[xi]
+ v = torch.zeros((len(label_id), nc + 5), device=x.device)
+ v[:, :4] = label_id[:, 1:5] # box
+ v[:, 4] = 1.0 # conf
+ v[range(len(label_id)), label_id[:, 0].long() + 5] = 1.0 # cls
+ x = torch.cat((x, v), 0)
+
+ # If none remain process next image
+ if not x.shape[0]:
+ continue
+
+ # Compute conf
+ x[:, 5:] *= x[:, 4:5] # conf = obj_conf * cls_conf
+
+ # Box (center x, center y, width, height) to (x1, y1, x2, y2)
+ box = xywh2xyxy(x[:, :4])
+
+ # Detections matrix nx6 (xyxy, conf, cls)
+ if multi_label:
+ i, j = (x[:, 5:] > conf_thres).nonzero(as_tuple=False).T
+ x = torch.cat((box[i], x[i, j + 5, None], j[:, None].float()), 1)
+ else: # best class only
+ conf, j = x[:, 5:].max(1, keepdim=True)
+ x = torch.cat((box, conf, j.float()), 1)[conf.view(-1) > conf_thres]
+
+ # Filter by class
+ if classes is not None:
+ x = x[(x[:, 5:6] == torch.tensor(classes, device=x.device)).any(1)]
+
+ # Check shape
+ n = x.shape[0] # number of boxes
+ if not n: # no boxes
+ continue
+
+ x = x[x[:, 4].argsort(descending=True)] # sort by confidence
+
+ # Batched NMS
+ c = x[:, 5:6] * (0 if agnostic else max_wh) # classes
+ boxes, scores = x[:, :4] + c, x[:, 4] # boxes (offset by class), scores
+ i = torchvision.ops.nms(boxes, scores, iou_thres) # NMS
+ if merge and (1 < n < 3e3): # Merge NMS (boxes merged using weighted mean)
+ # update boxes as boxes(i,4) = weights(i,n) * boxes(n,4)
+ iou = box_iou(boxes[i], boxes) > iou_thres # iou matrix
+ weights = iou * scores[None] # box weights
+ x[i, :4] = torch.mm(weights, x[:, :4]).float() / weights.sum(1, keepdim=True) # merged boxes
+ if redundant:
+ i = i[iou.sum(1) > 1] # require redundancy
+
+ output[xi] = x[i]
+ if (time.time() - t) > time_limit:
+ print(f"WARNING: NMS time limit {time_limit}s exceeded")
+ break # time limit exceeded
+
+ return output
+
+
+def scale_coords_landmarks(img1_shape, coords, img0_shape, ratio_pad=None):
+ # Rescale coords (xyxy) from img1_shape to img0_shape
+ if ratio_pad is None: # calculate from img0_shape
+ gain = min(img1_shape[0] / img0_shape[0], img1_shape[1] / img0_shape[1]) # gain = old / new
+ pad = (img1_shape[1] - img0_shape[1] * gain) / 2, (img1_shape[0] - img0_shape[0] * gain) / 2 # wh padding
+ else:
+ gain = ratio_pad[0][0]
+ pad = ratio_pad[1]
+
+ coords[:, [0, 2, 4, 6, 8]] -= pad[0] # x padding
+ coords[:, [1, 3, 5, 7, 9]] -= pad[1] # y padding
+ coords[:, :10] /= gain
+ coords[:, 0].clamp_(0, img0_shape[1]) # x1
+ coords[:, 1].clamp_(0, img0_shape[0]) # y1
+ coords[:, 2].clamp_(0, img0_shape[1]) # x2
+ coords[:, 3].clamp_(0, img0_shape[0]) # y2
+ coords[:, 4].clamp_(0, img0_shape[1]) # x3
+ coords[:, 5].clamp_(0, img0_shape[0]) # y3
+ coords[:, 6].clamp_(0, img0_shape[1]) # x4
+ coords[:, 7].clamp_(0, img0_shape[0]) # y4
+ coords[:, 8].clamp_(0, img0_shape[1]) # x5
+ coords[:, 9].clamp_(0, img0_shape[0]) # y5
+ return coords
diff --git a/r_facelib/detection/yolov5face/utils/torch_utils.py b/r_facelib/detection/yolov5face/utils/torch_utils.py new file mode 100644 index 0000000..f702962 --- /dev/null +++ b/r_facelib/detection/yolov5face/utils/torch_utils.py @@ -0,0 +1,40 @@ +import torch
+from torch import nn
+
+
+def fuse_conv_and_bn(conv, bn):
+ # Fuse convolution and batchnorm layers https://tehnokv.com/posts/fusing-batchnorm-and-conv/
+ fusedconv = (
+ nn.Conv2d(
+ conv.in_channels,
+ conv.out_channels,
+ kernel_size=conv.kernel_size,
+ stride=conv.stride,
+ padding=conv.padding,
+ groups=conv.groups,
+ bias=True,
+ )
+ .requires_grad_(False)
+ .to(conv.weight.device)
+ )
+
+ # prepare filters
+ w_conv = conv.weight.clone().view(conv.out_channels, -1)
+ w_bn = torch.diag(bn.weight.div(torch.sqrt(bn.eps + bn.running_var)))
+ fusedconv.weight.copy_(torch.mm(w_bn, w_conv).view(fusedconv.weight.size()))
+
+ # prepare spatial bias
+ b_conv = torch.zeros(conv.weight.size(0), device=conv.weight.device) if conv.bias is None else conv.bias
+ b_bn = bn.bias - bn.weight.mul(bn.running_mean).div(torch.sqrt(bn.running_var + bn.eps))
+ fusedconv.bias.copy_(torch.mm(w_bn, b_conv.reshape(-1, 1)).reshape(-1) + b_bn)
+
+ return fusedconv
+
+
+def copy_attr(a, b, include=(), exclude=()):
+ # Copy attributes from b to a, options to only include [...] and to exclude [...]
+ for k, v in b.__dict__.items():
+ if (include and k not in include) or k.startswith("_") or k in exclude:
+ continue
+
+ setattr(a, k, v)
|