search

yjmade / ios_camera_object_detection

Realtime mobile visualize based Object Detection based on TensorFlow and YOLO model
102 stars 23 forks source link

Xcode keeps disconnecting the app. #2

Open Shoshin23 opened 7 years ago

Shoshin23 commented 7 years ago

Hey @yjmade

Nice work on this. There are a couple of things that's sort of an issue for me. 1) Xcode keeps disconnecting the app and asking me to attach it to the process again. Is this some issue with the memory? I'm running this on an iphone 6s. 2) The inference is delayed a lot. It's not exactly realtime as it is when i run yolo on my webcam. Any reasons for this? thanks!

yjmade commented 7 years ago

for 1, no idea since I haven't work on it for a long time. for 2, do you enable optimizing when compiling tensorflow? with -O2, performance improved a lot.

Shoshin23 commented 7 years ago

I don't think I enabled optimising while compiling Tensorflow. I'll check it and let you know. On Fri, 5 May 2017 at 2:02 PM, Jay Young notifications@github.com wrote:

for 1, no idea since I haven't work on it for a long time. for 2, do you enable optimizing when compiling tensorflow? with -O2, performance improved a lot.

— You are receiving this because you authored the thread. Reply to this email directly, view it on GitHub https://github.com/yjmade/ios_camera_object_detection/issues/2#issuecomment-299410847, or mute the thread https://github.com/notifications/unsubscribe-auth/AB_0NStUlDdO--WTcjbuYMl_twuKSAGoks5r2t6igaJpZM4NRnlH .

BluehackRano commented 7 years ago

@Shoshin23 I've got same issue. screen_shot

And also got error logs below = = = = = = = = = = = = = = = = = = = : E /tensorflow/tensorflow/examples/ios/ios_camera_object_detection/CameraExampleViewController.mm:387] Running model failed:Invalid argument: Session was not created with a graph before Run()! = = = = = = = = = = = = = = = = = = =