github/DeepFaceLive
1
0
Fork 0
mirror of https://github.com/iperov/DeepFaceLive synced 2025-08-19 13:09:58 -07:00
Code Issues Projects Releases Packages Wiki Activity Actions

update xlib.facemeta

Browse source
This commit is contained in:
iperov 2021-10-22 19:32:21 +04:00
commit 57fc35cc3e
1 changed files with 16 additions and 5 deletions
Download patch file Download diff file Expand all files Collapse all files

21
xlib/facemeta/FLandmarks2D.py
View file

@ -101,8 +101,8 @@ class FLandmarks2D:
r = max(xrt[0], xrb[0]) r = max(xrt[0], xrb[0])
b = max(xlb[1], xrb[1]) b = max(xlb[1], xrb[1])
return FRect.from_ltrb( (l,t,r,b) ) return FRect.from_ltrb( (l,t,r,b) )
def calc_cut(self, w_h, coverage : float, output_size : int, def calc_cut(self, h_w, coverage : float, output_size : int,
exclude_moving_parts : bool, exclude_moving_parts : bool,
head_yaw : float = None, head_yaw : float = None,
x_offset : float = 0, y_offset : float = 0): x_offset : float = 0, y_offset : float = 0):
@ -114,8 +114,9 @@ class FLandmarks2D:
mat, matrix to transform img space to face_image space mat, matrix to transform img space to face_image space
uni_mat matrix to transform uniform img space to uniform face_image space uni_mat matrix to transform uniform img space to uniform face_image space
""" """
h,w = h_w
type = self._type type = self._type
lmrks = (self._ulmrks * w_h).astype(np.float32) lmrks = (self._ulmrks * (w,h)).astype(np.float32)
# estimate landmarks transform from global space to local aligned space with bounds [0..1] # estimate landmarks transform from global space to local aligned space with bounds [0..1]
if type == ELandmarks2D.L68: if type == ELandmarks2D.L68:
@ -161,7 +162,7 @@ class FLandmarks2D:
# calc affine transform from 3 global space points to 3 local space points size of 'output_size' # calc affine transform from 3 global space points to 3 local space points size of 'output_size'
mat = Affine2DMat.from_3_pairs ( l_t, np.float32(( (0,0),(output_size,0),(output_size,output_size) ))) mat = Affine2DMat.from_3_pairs ( l_t, np.float32(( (0,0),(output_size,0),(output_size,output_size) )))
uni_mat = Affine2DUniMat.from_3_pairs ( (l_t / w_h).astype(np.float32), np.float32(( (0,0),(1,0),(1,1) )) ) uni_mat = Affine2DUniMat.from_3_pairs ( (l_t / (w,h) ).astype(np.float32), np.float32(( (0,0),(1,0),(1,1) )) )
return mat, uni_mat return mat, uni_mat
@ -196,7 +197,7 @@ class FLandmarks2D:
""" """
h,w = img.shape[0:2] h,w = img.shape[0:2]
mat, uni_mat = self.calc_cut( (w,h), coverage, output_size, exclude_moving_parts, head_yaw=head_yaw, x_offset=x_offset, y_offset=y_offset) mat, uni_mat = self.calc_cut( (h,w), coverage, output_size, exclude_moving_parts, head_yaw=head_yaw, x_offset=x_offset, y_offset=y_offset)
face_image = cv2.warpAffine(img, mat, (output_size, output_size), cv2.INTER_CUBIC ) face_image = cv2.warpAffine(img, mat, (output_size, output_size), cv2.INTER_CUBIC )
return face_image, uni_mat return face_image, uni_mat
@ -213,6 +214,16 @@ class FLandmarks2D:
for x, y in pts: for x, y in pts:
cv2.circle(img, (x, y), radius, color, lineType=cv2.LINE_AA) cv2.circle(img, (x, y), radius, color, lineType=cv2.LINE_AA)
def get_convexhull_mask(self, h_w, color=(1,), dtype=np.float32) -> np.ndarray:
"""
"""
h, w = h_w
ch = len(color)
lmrks = (self._ulmrks * h_w).astype(np.int32)
mask = np.zeros( (h,w,ch), dtype=dtype)
cv2.fillConvexPoly( mask, cv2.convexHull(lmrks), color)
return mask