khazit / lip2word Goto Github PK
View Code? Open in Web Editor NEWEnd-to-end pipeline for lip reading at the word level using a tensorflow CNN implementation.
End-to-end pipeline for lip reading at the word level using a tensorflow CNN implementation.
Could you please send me link for downloading model.. I'm lack of dataset as well as hardware requirements (though I can use Google Colab but i don't know how to load such large dataset into Colab for training)
hey!
Can you send me the pre-train model....My laptop lacks the hardware requirements.
My email id - [email protected]
In capture_process_frames function, it will not be able to manage more than 29 frames, hence have a condition check or push it under try block! Check the Code below
while vidObj.isOpened():
success, image = vidObj.read()
if success :
image = cv2.cvtColor(image, cv2.COLOR_BGR2GRAY)
image = resize_frame(image, size_frame-180, offset=35)
image = cv2.resize(image, dsize=(size, size), interpolation=cv2.INTER_LINEAR)
try:
all_frames[countsize:size(count+1), :] = image
except:
break
count += 1
Hey! I really am interested in your code and was wondering if you could upload your pretrained model so that I can run some tests on it. My computer isn't really strong enough to train it from scratch.
Hoping for a quick reply,
Hello. I want to see the results of the pretrained model. Can you send it by email?
My E-mail is [email protected]
Thank you.
In frameAdjust function defination, please add else block and push the code as mentioned below:
if target == n_frames :
print("Perfect number of frames !")
return video
else :
if n_frames > target :
# If number of frames is more than 29, we select
# 29 evenly distributed frames
print("Adjusting number of frames")
idx = np.linspace(0, n_frames-1, 29)
idx = np.around(idx, 0).astype(np.int32)
print("Indexes of the selected frames : \n{}".format(idx))
return video[:, :, idx]
else :
# If number of frames is less than 29, duplicate last
# frame at the end of the video
output_video = np.zeros((video.shape[0], video.shape[1], 29)).astype(np.uint8)
output_video[:, :, :n_frames] = video
for i in range(target-n_frames+1) :
output_video[:, :, i+n_frames-1] = output_video[:, :, n_frames-1]
return output_video
Could you please send me link for downloading model.. I'm lack of dataset as well as hardware requirements.
A declarative, efficient, and flexible JavaScript library for building user interfaces.
๐ 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. ๐๐๐
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 โค๏ธ Open Source for everyone.
Alibaba Open Source for everyone
Data-Driven Documents codes.
China tencent open source team.