Hello,
I am Aqeel, a PhD candidate at Georgia Institute of Technology, working towards energy-effecient machine learning systems design.
Convert face dataset to masked dataset
Home Page: http://www.masktheface.aqeel-anwar.com
License: MIT License
this error shows up everytime 'random' category of mask_type argument is is passed. It is working for other types of masks, the error is for only 'surgical blue' mask type.
Can I use this approach with grayscale face images images?
If it is possible, could you share the script to download the real dataset from internet? Thanks
/home/Facial-Similarity-with-Siamese-Networks-in-Pytorch-master/facemask/utils/aux_functions.py:111: RankWarning: Polyfit may be poorly conditioned
perp_line = fit_line(x, y, image)
Nice work. If i want to extend the work for the glasses to the face. How could I do it? Thanks
Hi, I found there are some white edges of masks on faces. And how to eliminate this unexpected phenomenon? Thanks~
Great project, thank you! Is there a way to change the position of the mask on the face? Thanks in advance!
(python3.6) E:\face_mask\MaskTheFace>python mask_the_face.py --path 'data/frontal1.jpg' --mask_type N95 --verbose --write_original_image
Traceback (most recent call last):
File "mask_the_face.py", line 86, in <module>
args.predictor = dlib.shape_predictor(path_to_dlib_model)
RuntimeError: Error deserializing object of type int
Hello,
i tried to add mask to LFW but face cannot be detected for some pics. how can i solve this problem? is it possible to provide the LFW-SM datasets stated in the paper?
Thanks
Hi @aqeelanwar,
how to build the datasaet with mask to train arcface?
Best regards,
PeterPham
Hi @aqeelanwar,
Thanks for your great work!
Sometimes the faces are not detected from the dlib face detector. Could you please let me know if there is any place where I can set a lower face detection confidence value in the code?
I had the saw that in the detector(image, int for upsample)
. I believe, the second argument for upsampling would upsample the image by a factor and then try to detect the face, if it is so, I tried that but it did not work for a few low quality images.
Thanking you in anticipation.
Masking image file
Faces found: 6
Traceback (most recent call last):
File "mask_the_face.py", line 210, in <module>
cv2.imwrite(w_path, img)
cv2.error: OpenCV(4.2.0) /io/opencv/modules/imgcodecs/src/loadsave.cpp:661: error: (-2:Unspecified error) could not find a writer for the specified extension in function 'imwrite_'
If so, what is the best TPR can be reached on both LFW-SM and MFR2?
Hello everyone,
I was working with approx 17k images and i ran the mask_the_face.py script to mask all the faces , after the entire process I noticed that the output folder only contains 5k images...rest 12k went missing..!!
(All the images are faces)
Awaiting positive response...!
Thank You.!
Hi, I'm trying to figure out how do your software cut out bad pictures (ex. pictures where landmark predicion is bad), I've searched in your code but I can't find the part where you do it.
I understand that the software is for face masks but how would it perform if encountering sun glasses/glasses?
how to variation by args.color?
can someone help me on how to runs this model? I really need for our thesis project. I installed all of the requirements but im stuck on how to run it in windows
while running the code it say module not found error
No module name dlib
How can I create a default mask image without adding color?
thanks for sharing, but your code cannot run on gpu to add mask.
very appreciate if you can correct this.
i've already installed dlib gpu version correctly but the code cannot use it.... (no process was shown under nvidia-smi
)
Very look forward to your reply
@aqeelanwar @woomurf
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
The Web framework for perfectionists with deadlines.
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
We are working to build community through open source technology. NB: members must have two-factor auth.
Open source projects and samples from Microsoft.
Google ❤️ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.