Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Add DeepLab #22

Merged
merged 2 commits into from
Apr 7, 2019
Merged

Add DeepLab #22

merged 2 commits into from
Apr 7, 2019

Conversation

see--
Copy link
Contributor

@see-- see-- commented Mar 31, 2019

This pull request adds support for DeepLab on Android. I also modified the example app to compare the different tasks more easily.

result

@shaqian
Copy link
Owner

shaqian commented Apr 1, 2019

Thank you for the PR. Very cool!
Does DeepLab require the experimental GPU version of tensorflow-lite?

@see--
Copy link
Contributor Author

see-- commented Apr 1, 2019

Thank you for this cool repo! No, I didn't try the GPU version yet. The model is from the GPU example page https://www.tensorflow.org/lite/performance/gpu. It should therefore work with the GPU.

To improve the performance we could also reuse the ByteBuffers https://github.com/shaqian/flutter_tflite/blob/master/android/src/main/java/sq/flutter/tflite/TflitePlugin.java#L255. Just allocate one input and one output when we load the model and reuse them when we call tfLite.run. What do you think?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants