Professional Documents
Culture Documents
Learning Computer Vision
Learning Computer Vision
F. GAN
Technologies :
● GAN
important
Applications :
● autonomous cars
● autonomous drones
● solving captcha / OCR
● art
People to follow :
https://medium.com/@ageitgey/machine-learning-is-fun-
part-4-modern-face-recognition-with-deep-learning-
c3cffc121d78 with a full face
detection/alignment/recognition pipeline
Courses :
Related fields :
as input layer
Face detection
Face detection is about placing boxes around faces
algorithms to do that.
https://github.com/nodefluxio/face-detector-benchmark provide a
implementation code.
Haar classifiers
haar features
They are the old computer vision method present in opencv since
http://wearables.cc.gatech.edu/paper_of_week/viola01rapid.pdf.
for object detection. Haar classifiers are fast but have a low
accuracy.
https://docs.opencv.org/3.4.3/d7/d8b/tutorial_py_face_detection.ht
ml
gradients on the pixel of your images. These features are then fed
face_recognition (https://github.com/ageitgey/face_recognition)
lib.
MTCNN
https://kpzhang93.github.io/MTCNN_face_detection_alignment/in
dex.html
MobileNet
The best and fastest method these days for face detection. Based
https://arxiv.org/abs/1704.04861
Object detection
Object detection on many kind of objects
detection.
sensitive-score-maps-object-detection-91cd2389345c r-
fcn
● https://towardsdatascience.com/r-cnn-fast-r-cnn-faster-r-
cnn-yolo-object-detection-algorithms-36d53571365e a
A convolution layer
Object recognition
Object recognition is the general problem of classifying object into
The more successful neural networks have been using more and
more layer.
The ResNet architecture is the best to classify object to date.
Resnet architecture
That’s the reason why methods that don’t require retraining every
time on such big datasets are very useful. Transfer learning and
https://github.com/tensorflow/tensor2tensor#image-classification
Face recognition
Face recognition is about figuring out who is a face.
Historic methods
The historic way to solve that task has been to apply either
Facenet
person.
Triplet loss
The end result is each face (even faces not present in the original
of other people.
These embeddings can then be used with any machine learning
https://github.com/ageitgey/face_recognition
https://github.com/davidsandberg/facenet
https://hypraptive.github.io/2017/01/21/facenet-for-bears.html
Transfer learning
flowers.
https://medium.com/@14prakash/transfer-learning-using-keras-
Image segmentation
Image segmentation for autonomous driving
image.
mask-r-cnn-ebe6d793272
GAN
generator.
● The discriminator detects whether a picture is a class, it
dataset.
https://arxiv.org/abs/1809.11096
https://github.com/eriklindernoren/Keras-GAN
GPU yourself.
With as little as 1000$ it’s possible to build a decent machine to
https://hypraptive.github.io/2017/02/13/dl-computer-build.html
Vision in UI
Face dashboard of ownphotos
Ownphotos is an amazing UI allowing you to import your photos
It uses :