This paper evaluates the performance of a semi-supervised learning framework, with a Mean Teacher algorithm, used for semantic segmentation of pet images from Oxford-IIIT Pet dataset. In particular, we compare the framework results with a baseline without unlabelled data, and we also compare it with an upperbound with all labels being available. Furthermore, we tackle two research questions. The first question relates to the impact of different noise levels, added to the inputs, on the performance of the framework. The second question focuses on the impact of different labelled data ratios on the framework’s performance. Our experiments show that the framework performs better than the baseline by achieving 78.5% test IoU accuracy which is significantly higher than the baseline which gives 48.9%. The framework with an upper-bound performs the best as it gives 91.2% accuracy. The result of our research on the noise level shows that initially increasing the noise level improves test accuracy, but after a certain level it causes the accuracy to decrease. Lastly, we observe that increasing the labelled that ratio improves model accuracy only to a certain point, suggesting that not having enough labelled data is not the only factor which limits model accuracy, but other factors also play a part.
applied_dl's Introduction
applied_dl's People
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.