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

No detection box #3

Open
dahwang22 opened this issue Feb 3, 2018 · 5 comments
Open

No detection box #3

dahwang22 opened this issue Feb 3, 2018 · 5 comments

Comments

@dahwang22
Copy link

Hi, I had success run the swift project on my device, but there is no any detection box even I move my 6s around, can you help me fix it?

@csharpseattle
Copy link
Owner

It is possible the graph failed to load. Do you see any errors in the console?

@dahwang22
Copy link
Author

There is no any error message in the console

@csharpseattle
Copy link
Owner

If everything is running okay you should periodically see something like this in the console:
Avg. prediction time: 0.184770

To track down the problem you can try different lighting and use different phone orientations. The list of objects it should recognize is in the file mscoco_label_map.txt . Also, I recommend building with ssd_mobilenet_v1 (the default). The processing time on faster_rcnn and inception_v2 can be much longer.

@dahwang22
Copy link
Author

ok, thank you for your recommend, i will try it soon.

@csharpseattle
Copy link
Owner

Ok, let me know how it goes. Another thing you can try is to set the PredictionThreshold lower at the top of TensorflowGraph.mm. Bounding boxes are only drawn for percentages greater than this amount.

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

No branches or pull requests

2 participants