yumaofan / jetson_nano_demo

Some great implement of deep learning algorithm in Nvidia jetson nano platform. Such as face recognition, object detection, etc.

Hi there 👋

jetson_nano_demo's Issues

FPS for performance

i tested on my laptop with GTX 1050 and jetson nano too,it show that the FPS increase/decrease fastly depend on how many faces in image.
i wonder why FPS changes so fastly ( 37-68 on laptop, 3-11 on jetson nano)
i calculated FPS base on 0.95* last_FPS+ 0.05*current_FPS.

REport error when run camera_demo.py on jetson nano

jetson@jetson-desktop:~/Downloads/jetson_nano_demo-master/face_recognition$ sudo python3 camera_demo.py
2019-12-27 15:50:29.284513: I tensorflow/stream_executor/platform/default/dso_loader.cc:44] Successfully opened dynamic library libcudart.so.10.0
Traceback (most recent call last):
File "camera_demo.py", line 10, in
from utils import face_preprocess
File "/home/jetson/Downloads/jetson_nano_demo-master/face_recognition/utils/face_preprocess.py", line 4, in
from skimage import transform as trans
File "/usr/local/lib/python3.6/dist-packages/skimage/transform/init.py", line 2, in
from .hough_transform import (hough_line, hough_line_peaks,
File "/usr/local/lib/python3.6/dist-packages/skimage/transform/hough_transform.py", line 2, in
from ._hough_transform import (_hough_circle,
File "skimage/transform/_hough_transform.pyx", line 13, in init skimage.transform._hough_transform
File "/usr/local/lib/python3.6/dist-packages/skimage/draw/init.py", line 5, in
from .draw3d import ellipsoid, ellipsoid_stats
File "/usr/local/lib/python3.6/dist-packages/skimage/draw/draw3d.py", line 2, in
from scipy.special import (ellipkinc as ellip_F, ellipeinc as ellip_E)
File "/usr/lib/python3/dist-packages/scipy/special/init.py", line 648, in
from ._ellip_harm import ellip_harm, ellip_harm_2, ellip_normal
File "/usr/lib/python3/dist-packages/scipy/special/_ellip_harm.py", line 7, in
from ._ellip_harm_2 import _ellipsoid, _ellipsoid_norm
ImportError: cannot import name '_ellipsoid'

My pip3 list is following

etson@jetson-desktop:~/Downloads/jetson_nano_demo-master/face_recognition$ pip3 list
Package Version

---

absl-py 0.9.0
apt-clone 0.2.1
apturl 0.5.2
asn1crypto 0.24.0
astor 0.8.1
attrs 17.4.0
beautifulsoup4 4.6.0
blinker 1.4
Brlapi 0.6.6
cachetools 4.0.0
certifi 2018.1.18
chardet 3.0.4
Click 7.0
cryptography 2.1.4
cupshelpers 1.0
cycler 0.10.0
decorator 4.4.1
defer 1.0.6
distro-info 0.18ubuntu0.18.04.1
dlib 19.17.0
face-recognition 1.2.3
face-recognition-models 0.3.0
feedparser 5.2.1
gast 0.2.2
google-auth 1.10.0
google-auth-oauthlib 0.4.1
google-pasta 0.1.8
graphsurgeon 0.4.1
grpcio 1.26.0
h5py 2.10.0
html5lib 0.999999999
httplib2 0.9.2
idna 2.6
imageio 2.6.1
joblib 0.11
Keras-Applications 1.0.8
Keras-Preprocessing 1.1.0
keyring 10.6.0
keyrings.alt 3.0
language-selector 0.1
launchpadlib 1.10.6
lazr.restfulclient 0.13.5
lazr.uri 1.0.3
louis 3.5.0
lxml 4.2.1
macaroonbakery 1.1.3
Mako 1.0.7
Markdown 3.1.1
MarkupSafe 1.0
matplotlib 2.1.1
mock 3.0.5
networkx 2.4
nose 1.3.7
numpy 1.18.0
oauth 1.0.1
oauthlib 3.1.0
olefile 0.45.1
opt-einsum 3.1.0
PAM 0.4.2
pbr 5.4.4
Pillow 6.2.1
pip 19.3.1
pluggy 0.6.0
portpicker 1.3.1
protobuf 3.11.2
psutil 5.6.7
py 1.5.2
py-cpuinfo 5.0.0
pyasn1 0.4.8
pyasn1-modules 0.2.7
pycairo 1.16.2
pycrypto 2.6.1
pycups 1.9.73
pygobject 3.26.1
PyICU 1.9.8
PyJWT 1.5.3
pymacaroons 0.13.0
PyNaCl 1.1.2
pyparsing 2.2.0
pyRFC3339 1.0
pytest 3.3.2
python-apt 1.6.4
python-dateutil 2.6.1
python-debian 0.1.32
pytz 2018.3
PyWavelets 0.5.1
pyxdg 0.25
PyYAML 3.12
requests 2.22.0
requests-oauthlib 1.3.0
requests-unixsocket 0.1.5
rsa 4.0
scikit-image 0.16.2
scikit-learn 0.19.1
scipy 0.19.1
SecretStorage 2.3.1
setuptools 42.0.2
simplejson 3.13.2
six 1.13.0
ssh-import-id 5.7
system-service 0.3
systemd-python 234
tensorboard 2.0.2
tensorflow-estimator 2.0.1
tensorflow-gpu 2.0.0+nv19.12.tf2
tensorrt 5.1.6.1
termcolor 1.1.0
testresources 2.0.1
ubuntu-drivers-common 0.0.0
uff 0.6.3
unity-scope-calculator 0.1
unity-scope-chromiumbookmarks 0.1
unity-scope-colourlovers 0.1
unity-scope-devhelp 0.1
unity-scope-firefoxbookmarks 0.1
unity-scope-manpages 0.1
unity-scope-openclipart 0.1
unity-scope-texdoc 0.1
unity-scope-tomboy 0.1
unity-scope-virtualbox 0.1
unity-scope-yelp 0.1
unity-scope-zotero 0.1
urllib3 1.22
wadllib 1.3.2
webencodings 0.5
Werkzeug 0.16.0
wheel 0.33.6
wrapt 1.11.2
xkit 0.0.0
zope.interface 4.3.2

Can you help me?

Thanks!

Accuracy Improvement

Hi @yumaofan
you have done really amazing job. Solution is very robust and useful for demos.

But I was wondering if there's any future plan to change the embedding not only for single image but for multiple images of single person. It really helps in improving results further on.

What do you think ?

warped = cv2.warpAffine(img,M,(image_size[1],image_size[0]), borderValue = 0.0)报错

请问我一张图片中出现两张脸，我逐个对脸进行处理的时候报这个错误
cv2.error: OpenCV(4.1.1) /io/opencv/modules/imgproc/src/imgwarp.cpp:2594: error: (-215:Assertion failed) src.cols > 0 && src.rows > 0 in function 'warpAffine'
在处理第一个脸的时候不报错，直到第二个脸才会报错，这要怎么解决呢？
谢谢了

Performance was poor.

Unable to recognize faces. Any hint on how to improve? Thank you.

feature_compare

when you compare 2 embbeding by feature compare function,why using np.dot for 2 numpy array 1D,it seem return sum of ( element * element) of 2 numpy array
i dont understand why 2 numpy array similar it near 1 while they aren't similar feature_compare return value near 0

Crashes when it sees a face

As soon as someone comes near the camera, it crashes with this error :-

Traceback (most recent call last):
File "camera_demo.py", line 201, in
main()
File "camera_demo.py", line 181, in main
if dict[0][1] > VERIFICATION_THRESHOLD:
IndexError: list index out of range

Error when running on Jetson Nano

I am using Numpy 1.16.4.

Traceback (most recent call last):
  File "camera_demo.py", line 10, in <module>
    from utils import face_preprocess
  File "/home/yichengye/Desktop/jetson_nano_demo/face_recognition/utils/face_preprocess.py", line 4, in <module>
    from skimage import transform as trans
  File "/usr/lib/python3/dist-packages/skimage/__init__.py", line 158, in <module>
    from .util.dtype import *
  File "/usr/lib/python3/dist-packages/skimage/util/__init__.py", line 7, in <module>
    from .arraycrop import crop
  File "/usr/lib/python3/dist-packages/skimage/util/arraycrop.py", line 8, in <module>
    from numpy.lib.arraypad import _validate_lengths
ImportError: cannot import name '_validate_lengths'

仿射变换

我看到在MTCNN检测得到脸的坐标和关键点坐标之后，有一个仿射变换得到MobileFacenet输入的图像数据，请问可以直接根据MTCNN检测的脸的坐标来从原图上抠出脸的部分再resize为（112,112），送入MobileFacenet提取特征吗？仿射变换的意义是什么呢？

Object detection code

HI.
thanks for your work.

Is there any plan to add an object detection code into this repo?

report error when running the demo on nano board

hi,
when I run this demo on nano board, below error is reported:
which version is needed for numpy? my numpy version is 1.13.3.

$ python3 face_recognition/camera_demo.py
ModuleNotFoundError: No module named 'numpy.core._multiarray_umath'
ImportError: numpy.core.multiarray failed to import

The above exception was the direct cause of the following exception:

Traceback (most recent call last):
File "", line 968, in _find_and_load
SystemError: <class '_frozen_importlib._ModuleLockManager'> returned a result with an error set
ImportError: numpy.core._multiarray_umath failed to import
ImportError: numpy.core.umath failed to import
2019-07-28 21:59:25.983800: F tensorflow/python/lib/core/bfloat16.cc:675] Check failed: PyBfloat16_Type.tp_base != nullptr
Aborted (core dumped)

Real time add faces

Hi, is it possible to add unknown faces into face_db in real time?

Thank you!