Comments (5)
The accuracy for MobileNet is pretty much the same as VGG16. But it's 30 times smaller and about 10 times faster. So I'd definitely use MobileNet over VGG16.
from mobilenet-coreml.
Amazing, Definitely noticed the performance increase!
I just found your blog article about Real time object detection using Yolo, and I have to say it's brilliant and exactly what I was trying to achieve. (currently able to do 1 classification / frame @60fps using Metal and MobileNet, but I want to detect multiple objects and also know where they were detected in the frame... so I think I NEED to use YOLO.. )
I'd like to know what are the limitations of using YOLO, for example how does the TinyYOLO model compare to MobileNet, and how would I go about expanding your TinyYOLO model to be able to classify more different categories, or maybe use another model
EDIT: Which TinyYOLO model did you use to convert to .mlmodel? The VOC2007+2012 one or the COCO one? I'm very interested in the COCO one since it seems to be able to classify more categories (80 vs 20 of VOC) and it is a more recent challenge. Guess I will try to follow your conversion steps, but might be too much for my level of understanding (at the moment.. ;)
Thanks in advance!
Oh and BTW, I just noticed you're also from the Netherlands ;) I live in Utrecht myself.
from mobilenet-coreml.
I converted the VOC one.
Note that in the TensorFlow models repo is a version of SSD that runs on MobileNets. This is roughly as accurate as Tiny YOLO but runs much faster. (It's a fair bit of work to do the conversion, which I did for a client and therefore cannot share, but definitely worth it.)
from mobilenet-coreml.
@hollance You can not share the converted model but Can you please post some guidelines which we can follow to convert MobileNet SSD for the CoreML? This would be a great Help to Community. And atleast you can post some links which you have come across for getting this issue solved.
Btw, I am trying to convert TF SSD Mobilenet to CoreML but I am facing bit of a problem in finding right tools to be used.
from mobilenet-coreml.
I haven't converted MobileNet+SSD to Core ML, so there may be issues I don't know about but one issue is that the model is in TensorFlow format so you have to write your own converter. Also, you have to replave the relu6 activations with something else, as Core ML does not support it.
from mobilenet-coreml.
Related Issues (11)
- Where is synset_words.txt HOT 3
- Model looks unoptimised HOT 1
- Getting the output from intermediate layers in the mobilenetv2 network HOT 2
- build succeed but crash HOT 14
- Error: When trying to convert Caffe model to CoreML model HOT 3
- How long does it take to predict an image HOT 4
- Model not working for me HOT 3
- Why is Keras necessary? HOT 1
- Get wrong result after converting the model to coreML HOT 6
- [Question] Where are the conversion parameters come from? HOT 2
Recommend Projects
-
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
-
Vue.js
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
-
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
-
TensorFlow
An Open Source Machine Learning Framework for Everyone
-
Django
The Web framework for perfectionists with deadlines.
-
Laravel
A PHP framework for web artisans
-
D3
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
-
Recommend Topics
-
javascript
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
-
web
Some thing interesting about web. New door for the world.
-
server
A server is a program made to process requests and deliver data to clients.
-
Machine learning
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
-
Visualization
Some thing interesting about visualization, use data art
-
Game
Some thing interesting about game, make everyone happy.
Recommend Org
-
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
-
Microsoft
Open source projects and samples from Microsoft.
-
Google
Google ❤️ Open Source for everyone.
-
Alibaba
Alibaba Open Source for everyone
-
D3
Data-Driven Documents codes.
-
Tencent
China tencent open source team.
from mobilenet-coreml.