Re: Detecting face features using CoreImage or HSV space/Mask
Re: Detecting face features using CoreImage or HSV space/Mask
- Subject: Re: Detecting face features using CoreImage or HSV space/Mask
- From: Muhammad Hussein Nasrollahpour <email@hidden>
- Date: Fri, 02 Jan 2015 12:48:26 -0800
You hit the point. Yup, implementing my own algorithm was one of my idea BUT because I need to code it asap(in a weekend) I need to have a precise and fast(in terms of coding) solution to the problem.
What’s the best option to accomplish the task?
There is another solution to the problem too. We can let user to specify the position of eyes, lips, etc.
[muhammad says: @“Happy trails to you.”];
> On Jan 2, 2015, at 11:55 AM, Jens Alfke <email@hidden> wrote:
>
>
>> On Jan 2, 2015, at 12:34 AM, Muhammad Hussein Nasrollahpour <email@hidden <mailto:email@hidden>> wrote:
>>
>> I just got one solution to the problem. Instead of using CI I can convert RGB to HSV and then create a mask for face features e.g. lips, eyes, etc. but
>> 1. How to use hue to create a mask for different face features?
>
> Are you talking about implementing your own algorithm for facial feature detection? There's a whole lot more to it than just converting the color space! It's fairly advanced stuff — it's been a hot topic in artificial-intelligence research since the 1960s, and only became commercially feasible within the last 10 years due to fast CPUs and GPUs.
>
> You can probably search for research papers on facial analysis and feature detection (Google Scholar is a good way to find them) but be aware that this is not the kind of thing you can do in a weekend; it's more of a grad-school thesis project.
>
> —Jens
_______________________________________________
Cocoa-dev mailing list (email@hidden)
Please do not post admin requests or moderator comments to the list.
Contact the moderators at cocoa-dev-admins(at)lists.apple.com
Help/Unsubscribe/Update your Subscription:
This email sent to email@hidden