Native Camera/Corona Camera Fill Conflict Help!

Hi everyone,

Finally rockin’ and rollin’ with my Corona OpenCV Augmented Reality app!

The UI work is done in Corona, and the central element is a full screen camera fill like Snapchat.

Problem is, I’m using openCVs VideoCamera (for the reason of asynchronous mat manipulation) and it conflicts with the fill property Corona side. Either openCV shuts off the fill, or the fill shuts off openCV.

I need a work around, as I can’t use both at the same time and the VideoCamera object overlays Corona objects. Please help me out, I’m so so close.

Thank you,

Aidan

I’m not sure that’s going to work. I don’t think two processes can control the camera at the same time. 

Rob

Hi Rob,

Seems so…

Decent hack is to save the frames from VideoCamera and rapidly display them as an image fill in Corona.

If the Corona Labs team could dream up a crazy way to get native objects layered into the Corona canvas effectively, I don’t think I’d be the only happy one.

Thank you,

Aidan

Hello,

I like to create an app for iOS and Android with face detection functionality. 

Do you know if this is easy to combine the OpenCV library with Corona?

There are not many cross platform platforms who support face detection

functionality for both iOS and Android.

Best regards,

Marc

Hi Dikkesnoek,

OpenCV is hard to integrate into iOS. The problem is that iOS has been updated more often than OpenCV, so much so that most if not all OpenCV iOS documentation is wrong. The other big problem is trying to use a demo project any earlier than 6 months ago, as it will crash for upwards of 2 dozen individual reasons.

I’m well on my way with OpenCV now, though one of the first projects I tried out was a Corona + Face Detection. Do you need to identify where the face is on screen in realtime, or are you looking to do something more complicated?

Haven’t tried Android OpenCV yet, but I hear it’s super easy.

First step is to buy Corona Enterprise, then you can get started with OpenCV :slight_smile:

Regards,

Aidan

I don’t know why you couldn’t get OpenCV working with Corona Enterprise. OpenCV seems like a pretty broad scoped tool. It does a lot and it’s not something that can easily be made into a plugin. This will be something you will need Enterprise for.

Aidan seems to have the most experience with it.

Hi, thanks to all for your reply,

I only want to do some simple face detection. From a picture I

want to cut out the face only and save or send it as a picture.

I want to go Enterprise but I need to be sure that I can use

these kind of techniques to build an app running on iOS and

Android. In the past I’ve created a simple face detection app

running on iOS. With Core Image it is quiet easy to do. I don’t

have experience with OpenCV. I thought that this could be the

most easy way to do cross platform face detection in Corona

Labs.

Best regards,

Marc

You can get a trial subscription to Enterprise to see if it will do what you want.

Hi Rob,

Thanks. I will let you know. Now I need to convince my boss and collegue

to switch over to Corona. I think that this is not a big issue.

Best regards,

Marc

I’m not sure that’s going to work. I don’t think two processes can control the camera at the same time. 

Rob

Hi Rob,

Seems so…

Decent hack is to save the frames from VideoCamera and rapidly display them as an image fill in Corona.

If the Corona Labs team could dream up a crazy way to get native objects layered into the Corona canvas effectively, I don’t think I’d be the only happy one.

Thank you,

Aidan

Hello,

I like to create an app for iOS and Android with face detection functionality. 

Do you know if this is easy to combine the OpenCV library with Corona?

There are not many cross platform platforms who support face detection

functionality for both iOS and Android.

Best regards,

Marc

Hi Dikkesnoek,

OpenCV is hard to integrate into iOS. The problem is that iOS has been updated more often than OpenCV, so much so that most if not all OpenCV iOS documentation is wrong. The other big problem is trying to use a demo project any earlier than 6 months ago, as it will crash for upwards of 2 dozen individual reasons.

I’m well on my way with OpenCV now, though one of the first projects I tried out was a Corona + Face Detection. Do you need to identify where the face is on screen in realtime, or are you looking to do something more complicated?

Haven’t tried Android OpenCV yet, but I hear it’s super easy.

First step is to buy Corona Enterprise, then you can get started with OpenCV :slight_smile:

Regards,

Aidan

I don’t know why you couldn’t get OpenCV working with Corona Enterprise. OpenCV seems like a pretty broad scoped tool. It does a lot and it’s not something that can easily be made into a plugin. This will be something you will need Enterprise for.

Aidan seems to have the most experience with it.

Hi, thanks to all for your reply,

I only want to do some simple face detection. From a picture I

want to cut out the face only and save or send it as a picture.

I want to go Enterprise but I need to be sure that I can use

these kind of techniques to build an app running on iOS and

Android. In the past I’ve created a simple face detection app

running on iOS. With Core Image it is quiet easy to do. I don’t

have experience with OpenCV. I thought that this could be the

most easy way to do cross platform face detection in Corona

Labs.

Best regards,

Marc

You can get a trial subscription to Enterprise to see if it will do what you want.

Hi Rob,

Thanks. I will let you know. Now I need to convince my boss and collegue

to switch over to Corona. I think that this is not a big issue.

Best regards,

Marc