>|h(6ddlZddlZddlmZddlmZddlZddlm Z ddl m Z ddl ddl m Z Gd d eZGd d eZd dddZdddddddddZGddZy)N)IntEnum)io)version)tqdm)*) FolderDataceZdZdZdZdZdZy) LandmarksTypeaAEnum class defining the type of landmarks to detect. ``TWO_D`` - the detected points ``(x,y)`` are detected in a 2D space and follow the visible contour of the face ``TWO_HALF_D`` - this points represent the projection of the 3D points into 3D ``THREE_D`` - detect the points ``(x,y,z)``` in a 3D space rN)__name__ __module__ __qualname____doc__TWO_D TWO_HALF_DTHREE_DQ/var/www/html/test/engine/venv/lib/python3.12/site-packages/face_alignment/api.pyr r s EJGrr ceZdZdZy) NetworkSizeN)rrrLARGErrrrrs  ErrzFhttps://www.adrianbulat.com/downloads/python-fan/2DFAN4-cd938726ad.zipzFhttps://www.adrianbulat.com/downloads/python-fan/3DFAN4-4a694010b9.zipzEhttps://www.adrianbulat.com/downloads/python-fan/depth-6c4283c0e0.zip)z2DFAN-4z3DFAN-4depthzJhttps://www.adrianbulat.com/downloads/python-fan/2DFAN4_1.6-c827573f02.zipzJhttps://www.adrianbulat.com/downloads/python-fan/3DFAN4_1.6-ec5cf40a1d.zipzIhttps://www.adrianbulat.com/downloads/python-fan/depth_1.6-2aa3f18772.zipzJhttps://www.adrianbulat.com/downloads/python-fan/2DFAN4_1.5-a60332318a.zipzJhttps://www.adrianbulat.com/downloads/python-fan/3DFAN4_1.5-176570af4d.zipzIhttps://www.adrianbulat.com/downloads/python-fan/depth_1.5-bc10f98e39.zip)z1.6z1.5ceZdZejdej ddddfdZd dZej d dZ ej d dZ d d gd d ddfd Z y) FaceAlignmentcudaFsfdNc ||_||_||_||_||_t j tjt j dkr tdt|}tj} d| vr| jddd} n| jddd} d|vrd tjj_td |zt!t#|gd} |xsi}| j$d||d ||_|t(j*k(rd t-|z} nd t-|z} tj.j1t3t4j7| t8| |_|j:j=|||j:j?|t(j@k(rtj.j1t3t4j7| t8d|_!|jBj=|||jBj?yy)Nz1.5.0zUnsupported pytorch version detected. Minimum supported version of pytorch: 1.5.0 Either upgrade (recommended) your pytorch setup, or downgrade to face-alignment 1.2.0dev.r rrrTzface_alignment.detection.)deviceverbosez2DFAN-z3DFAN-dtyperr)"r$ flip_inputlandmarks_typer%r'rparsetorch __version__ ImportErrorintrsplitbackendscudnn benchmark __import__globalslocals FaceDetector face_detectorr rstrjitloadload_file_from_url models_urlsgetdefault_model_urlsface_alignment_nettoevalrdepth_prediciton_net) selfr) network_sizer$r'r(r7face_detector_kwargsr%pytorch_versionface_detector_module network_names r__init__zFaceAlignment.__init__6s $,  ==** +gmmG.D D!rt t<( ++ O #-44S!1>>vfV]vauv ]00 0#c,&77L#c,&77L"')).. {@RST`a b#d ""6"7 $$& ]22 2(- ";???DV#WX_#`a)cD %  % % ( (u ( =  % % * * , 3rc*|j||||S)a4Deprecated, please use get_landmarks_from_image Arguments: image_or_path {string or numpy.array or torch.tensor} -- The input image or path to it Keyword Arguments: detected_faces {list of numpy.array} -- list of bounding boxes, one for each face found in the image (default: {None}) return_bboxes {boolean} -- If True, return the face bounding boxes in addition to the keypoints. return_landmark_score {boolean} -- If True, return the keypoint scores along with the keypoints. )get_landmarks_from_image)rC image_or_pathdetected_faces return_bboxesreturn_landmark_scores r get_landmarkszFaceAlignment.get_landmarkses,,]NM[pqqrc \t|}|)|jj|j}t |dk(rt j d|s|ryyg}g}t|D]\}} tj| d| d| dz dz z | d| d| dz dz z g} | d| d| dz d zz | d<| d| dz | dz| dz |jjz } t|| | } tj| jd j} | j|j |j" } | j%d j'd|j)| j+} |j,r6| t/|j)t/| j+d z } | jdtj0j3} t5| | j3| \}}}tj|tj|}}|j7dddz|j7dd}}|j9d}|j:t<j>k(r=tAjBdt@j0 }tEdD].}||dfdkDs||dfdkDstG||||d||<0tj|j'd}|j|j |j" }|jItjJ| |fdjLjOj7ddjtj0 }tjJ||ddd| zz z zfd}|jQ|j3|jQ||sd}|sd}|s|r|||fS|S)aPredict the landmarks for each face present in the image. This function predicts a set of 68 2D or 3D images, one for each image present. If detect_faces is None the method will also run a face detector. Arguments: image_or_path {string or numpy.array or torch.tensor} -- The input image or path to it. Keyword Arguments: detected_faces {list of numpy.array} -- list of bounding boxes, one for each face found in the image (default: {None}) return_bboxes {boolean} -- If True, return the face bounding boxes in addition to the keypoints. return_landmark_score {boolean} -- If True, return the keypoint scores along with the keypoints. Return: result: 1. if both return_bboxes and return_landmark_score are False, result will be: landmark 2. Otherwise, result will be one of the following, depending on the actual value of return_* arguments. (landmark, landmark_score, detected_face) (landmark, None, detected_face) (landmark, landmark_score, None ) NrNo faces were detected.NNNr g@r rgQ?)r rrr&go@T)is_labelcpu)r$r'Dr)rVrWg?gp@gi@)) get_imager7detect_from_imagecopylenwarningswarn enumerater+tensorreference_scalecrop from_numpy transposefloatr@r$r'div_ unsqueeze_r?detachr(flipfloat32numpyget_preds_fromhmviewsqueezer)r rnpzerosrange draw_gaussianrBcatdatarUappend)rCrLrMrNrOimage landmarkslandmarks_scoresidcenterscaleinpoutptspts_imgscoresheatmaps depth_preds rrKz&FaceAlignment.get_landmarks_from_imagess4-(  !!//AA%**,ON ~ ! # MM3 4 5' n-' ,DAq\\11!++QqTQqTAaD[C4G-GHJFq QqTAaD[D$88F1IqTAaD[1Q4'!A$.$2D2D2T2TTEufe,C""3==$!EG &&DJJ&7C HHUO & &q )))#.557CtD33DI>EEGRVWW&&U]]&;AACC#3C#O C& ++C0%2B2B72KC88B?Q. R0CC^^A&F""m&;&;;88N"**Er4A1a4y1}QTQ&3$QKQ'4 4!++(jm$;;t{{$**;E!66IIsHoq13374dd2qk""SXS`S`"Ja))jC5EEM3J,KLMqR   W]]_ -  # #F +O' ,R!N$#  1.> > rc||jj|}t|dk(rtjd|s|ryyg}g}t |D]\}}|j ||jjjddd||} |r| \} } } |j| n| } | tj| d} ng} |j| |sd}|sd}|s|r|||fS|S) aPredict the landmarks for each face present in the image. This function predicts a set of 68 2D or 3D images, one for each image in a batch in parallel. If detect_faces is None the method will also run a face detector. Arguments: image_batch {torch.tensor} -- The input images batch Keyword Arguments: detected_faces {list of numpy.array} -- list of bounding boxes, one for each face found in the image (default: {None}) return_bboxes {boolean} -- If True, return the face bounding boxes in addition to the keypoints. return_landmark_score {boolean} -- If True, return the keypoint scores along with the keypoints. Return: result: 1. if both return_bboxes and return_landmark_score are False, result will be: landmarks 2. Otherwise, result will be one of the following, depending on the actual value of return_* arguments. (landmark, landmark_score, detected_face) (landmark, None, detected_face) (landmark, landmark_score, None ) NrrRrSrr )rMrO)axis) r7detect_from_batchr[r\r]r^rKrUrjrcrtrn concatenate) rC image_batchrMrNrOrvlandmarks_scores_listrxfacesres landmark_setrw_s rget_landmarks_from_batchz&FaceAlignment.get_landmarks_from_batchs(6  !!//AA+NN ~ ! # MM3 4 5' "!.1 +HAu//A""$**,66q!Q?$&;0C %471 .%,,-=>" '!~~lC !   \ *! +$!N$$( ! 13^C C rz.jpgz.pngTct||jj|||j}tj j j|dddd}i} t|| D]g\} } | d| d} } |jj| } |s|r"|j| | ||\} }}| ||f| | <Q|j| | } | | | <i| S) aScan a directory for images with a given extension type(s) and predict the landmarks for each face present in the images found. Arguments: path {str} -- path to the target directory containing the images Keyword Arguments: extensions {list of str} -- list containing the image extensions considered (default: ['.jpg', '.png']) recursive {boolean} -- If True, scans for images recursively (default: True) show_progress_bar {boolean} -- If True displays a progress bar (default: True) return_bboxes {boolean} -- If True, return the face bounding boxes in addition to the keypoints. return_landmark_score {boolean} -- If True, return the keypoint scores along with the keypoints. rFr r) batch_sizeshuffle num_workersprefetch_factor)disabler)rNrO) r r7tensor_or_path_to_ndarrayr%r+utilsrs DataLoaderrrYrK)rCpath extensions recursiveshow_progress_barrNrOdataset dataloader predictions image_pathrubounding_boxespredsbboxscores rget_landmarks_from_directoryz*FaceAlignment.get_landmarks_from_directorysT4#5#5#O#OQ[]fhlhthtu[[%%00QPUcdvw0x  #' @Q^s&C&u"tU+0$*> J'55e^L*/ J' 0r)NFF) rrrrrr+rirIrPno_gradrKrrrrrrr5s4?4E4EemmUZqu@E--^ rU]]_Y^7<WWrU]]_W\7<@@D>DVr<rrrrrs # G 'XW T`_\  `_\   yyr