...
Код: Выделить всё
olej@R420:~/2023/DlibC++/step-3_face_recognition_aplit$ ./face_encoding.py 00000001.png
<class 'numpy.ndarray'>
1 <class 'list'>
<class 'tuple'>
(180, 469, 366, 283)
1 <class 'list'>
9 <class 'dict'>
{'chin': [(290, 240), (293, 263), (298, 285), (305, 307), (314, 326), (326, 343), (343, 357), (362, 367), (383, 369), (404, 364), (422, 351), (438, 336), (448, 316), (454, 294), (457, 271), (459, 248), (459, 224)], 'left_eyebrow': [(299, 215), (307, 201), (322, 193), (338, 194), (354, 200)], 'right_eyebrow': [(388, 197), (402, 189), (419, 187), (435, 192), (446, 203)], 'nose_bridge': [(372, 223), (373, 241), (373, 259), (374, 277)], 'nose_tip': [(357, 288), (366, 291), (376, 293), (386, 290), (394, 286)], 'left_eye': [(316, 236), (326, 231), (337, 230), (349, 234), (338, 237), (326, 238)], 'right_eye': [(398, 230), (408, 224), (419, 223), (430, 227), (420, 231), (409, 231)], 'top_lip': [(346, 324), (357, 318), (368, 314), (377, 315), (386, 312), (398, 314), (412, 318), (406, 318), (387, 320), (378, 322), (369, 322), (352, 324)], 'bottom_lip': [(412, 318), (401, 328), (389, 334), (380, 336), (371, 336), (359, 333), (346, 324), (352, 324), (370, 323), (379, 323), (387, 321), (406, 318)]}
chin [(290, 240), (293, 263), (298, 285), (305, 307), (314, 326), (326, 343), (343, 357), (362, 367), (383, 369), (404, 364), (422, 351), (438, 336), (448, 316), (454, 294), (457, 271), (459, 248), (459, 224)]
left_eyebrow [(299, 215), (307, 201), (322, 193), (338, 194), (354, 200)]
right_eyebrow [(388, 197), (402, 189), (419, 187), (435, 192), (446, 203)]
nose_bridge [(372, 223), (373, 241), (373, 259), (374, 277)]
nose_tip [(357, 288), (366, 291), (376, 293), (386, 290), (394, 286)]
left_eye [(316, 236), (326, 231), (337, 230), (349, 234), (338, 237), (326, 238)]
right_eye [(398, 230), (408, 224), (419, 223), (430, 227), (420, 231), (409, 231)]
top_lip [(346, 324), (357, 318), (368, 314), (377, 315), (386, 312), (398, 314), (412, 318), (406, 318), (387, 320), (378, 322), (369, 322), (352, 324)]
bottom_lip [(412, 318), (401, 328), (389, 334), (380, 336), (371, 336), (359, 333), (346, 324), (352, 324), (370, 323), (379, 323), (387, 321), (406, 318)]
1 <class 'list'>
128 <class 'numpy.ndarray'>
[-6.99085146e-02 9.84077826e-02 1.21112026e-01 3.21743079e-04
-8.06934685e-02 -3.00324205e-02 -2.96985395e-02 -6.87163696e-02
1.80488706e-01 -6.75299838e-02 1.58340216e-01 -2.03328710e-02
-2.56850660e-01 6.75959513e-02 -1.02667302e-01 1.90818638e-01
-1.33553475e-01 -1.25300393e-01 -7.63143077e-02 5.06927259e-04
-9.69897769e-03 5.19758426e-02 1.30818048e-02 5.43041490e-02
-8.44295695e-02 -4.11997288e-01 -9.06100795e-02 -9.69127864e-02
4.06597964e-02 -8.26481655e-02 -6.94197416e-02 1.06586754e-01
-1.10932887e-01 -4.59331349e-02 1.27042860e-01 3.46936658e-02
-2.30356120e-02 -7.51237497e-02 2.95453608e-01 7.30510131e-02
-2.49845490e-01 -3.19938548e-02 1.23676635e-01 3.10146213e-01
2.39223480e-01 4.04311307e-02 6.09315261e-02 -1.13170289e-01
1.79889232e-01 -2.66186446e-01 -3.27013731e-02 1.57478407e-01
8.86571333e-02 1.04656339e-01 8.42583850e-02 -1.68988556e-01
5.63557446e-03 4.32378277e-02 -2.05842003e-01 2.13440172e-02
9.83658899e-03 -7.29783475e-02 -1.60745829e-02 -1.90627649e-02
2.16138527e-01 2.32765730e-02 -1.32712394e-01 -9.71577018e-02
2.83064336e-01 -2.04470009e-01 1.60779916e-02 6.59383014e-02
-1.09041549e-01 -1.50104269e-01 -2.32336462e-01 -8.21434408e-02
4.95616853e-01 1.19069926e-01 -1.26885980e-01 7.11885318e-02
-1.34624496e-01 -1.24624029e-01 -2.27493364e-02 1.45198688e-01
-4.43203375e-02 9.13855899e-03 6.91940356e-03 1.47860209e-02
2.73463368e-01 -1.44846039e-02 5.05373329e-02 2.02544555e-01
3.26371305e-02 4.51780595e-02 2.32620891e-02 8.14667195e-02
-1.36481047e-01 -5.39681017e-02 -1.68556228e-01 -4.73946184e-02
-8.87599662e-02 -4.70957309e-02 5.66911101e-02 1.47512019e-01
-2.95983404e-01 1.58027336e-01 2.28147116e-02 -4.56366837e-02
1.88251138e-02 7.08771646e-02 -4.78914380e-02 -6.09939769e-02
1.54033735e-01 -2.81089604e-01 1.26432493e-01 1.87709451e-01
-4.99954866e-03 9.15239006e-02 -1.44863576e-02 3.22556794e-02
3.12945880e-02 -6.52948618e-02 -1.51842445e-01 -1.17168240e-01
2.14236649e-03 -4.57915142e-02 -2.86754295e-02 -8.76666512e-04]
Вот этого числового набора вполне достаточно чтобы распознавать этого человека в любом контексте.