Sidebar Sidebar
Become a MacRumors Supporter for $50/year with no ads, ability to filter front page stories, and private forums.

Looking for feedback for my new API

S

Starwars

macrumors newbie
Original poster
Mar 10, 2021
1
0
Hello,
I am building an API that let's you remove background of any image. You can process upto 500 images in one batch with the API. I am looking for feedback from this community.
Are you willing to try it for free (limited batch size) and give me feedback? Also, would you pay for something like this in the future?

Cheers.
 
This isn't something that'd interest me but I'm curious about your business use case. In what scenarios would someone have 500 images where they'd need to remove the background entirely? I would think that in even most "high shot count" scenarios like events and weddings, the context that a background provides would still be wanted.
 
I also would not have any use for this.
Even the very best photo editing programs almost never get the masking quite right. It doesn't take very long to do it myself, and I've never seen the need for batching such efforts. I edit my images one at a time.
I agree with r.harris1 that 1) it isn't clear to me what your target audience would be, and 2) the importance of the existing background as part of the scene or as an intentional backdrop.
We wish you good fortune with the project.
 
Starwars said:
Are you willing to try it for free (limited batch size) and give me feedback? Also, would you pay for something like this in the future?
Click to expand...

If you want good feedback from knowledgeable testers then don't give them a limited version. It's bad form, and you will not benefit from them getting full use of the program. I think you'd want to know if the API handles 50 or 100 images just fine but craps out at 250.
 
I would think the target would be someone who shoots portraits, like senior portraits or corporate, and would have a couple hundred photos that they could then replace the background on. Unfortunately, that’s not me so I can’t help out! But there’s already software doing that which exists (remove.bg), so I’m not sure how it compares to yours! There are also older offline software for that, especially if you’re using a solid background/green screen.
 
mackmgg said:
I would think the target would be someone who shoots portraits, like senior portraits or corporate, and would have a couple hundred photos that they could then replace the background on. Unfortunately, that’s not me so I can’t help out! But there’s already software doing that which exists (remove.bg), so I’m not sure how it compares to yours! There are also older offline software for that, especially if you’re using a solid background/green screen.
Click to expand...

I understand your reasoning but it does not match the real world. A friend of mine is a professional and shoots thousands of portraits and dozens of events throughout the year. He shows up with the right background or a green screen depending on the job. He would never spend the time to replace what he can do right in the first place.
 
deep diver said:
I understand your reasoning but it does not match the real world. A friend of mine is a professional and shoots thousands of portraits and dozens of events throughout the year. He shows up with the right background or a green screen depending on the job. He would never spend the time to replace what he can do right in the first place.
Click to expand...
Well yeah, I think a green screen would be the better use case anyway, since it would be more accurate. Would it not speed up workflows to have a batch method of removing the green screen? Of course you could probably do this with a Photoshop macro pretty easily too.
 
mackmgg said:
Well yeah, I think a green screen would be the better use case anyway, since it would be more accurate. Would it not speed up workflows to have a batch method of removing the green screen? Of course you could probably do this with a Photoshop macro pretty easily too.
Click to expand...

No. The backgrounds are already in the computer. The green screen software automatically puts in the background as an overlay on top of the green before anything else happens. It is a program driven additive process, not subtractive as OP wants to do.

I'd still like to hear from OP what kind of situation he envisions that would make this a useful tool.
 
Last edited:
Could you share some example pics? Would give all of us a better understanding of what you are talking about. Is this a standalone program? PS plug-in? LR plug-in?
 
This is a picture of Mrs. Diver and me. It happens that we were at a party for this friend's daughter. We were standing in front of a plain green screen. He uploaded the photo of us to the computer which put in the image of the Rock and Roll Hall of Fame wherever it saw green, and then sent it to the printer. Had I been wearing a light green suit, you would see the Hall of Fame there as well.

Edit: I just noticed OP's handle is Starwars. Green screens were used a lot in making those movies.
Sorry folks, but all of those exciting battle scenes in space were NOT real.
Sorry.
Take it up with your therapists.

IMG_20210317_204312 resize.jpg
 
Last edited:
mackmgg said:
Well yeah, I think a green screen would be the better use case anyway, since it would be more accurate. Would it not speed up workflows to have a batch method of removing the green screen? Of course you could probably do this with a Photoshop macro pretty easily too.
Click to expand...

I think I just realized what you are thinking. Tell me if I'm right. In PS you would cut out what you want to keep and then place that on top of the background you really want. That might be the way we would do a small number of images. A professional would also do that for particular reasons such as adding a single person to a photo. That is a very time intensive (and potentially expensive) project to get it to look real. I don't imagine a pro is going to do the cut and paste approach on such a large scale, especially when what I described is easily available and not very expensive. It is old technology. I know my friend wouldn't use the cut and paste approach except in those particular cases. As I said, he goes with the right tools or to the right place from the start.

Or, we could be using different language to describe the same thing. (That seems most likely.). Or, perhaps, except for those cut and paste moments, the language we're using is all wrong because it's being done in the computer under the control of THE ALGORITHM.

In either case, I'm not sure there would be a market for what OP is working on.
 
Last edited:
I
MacNut said:
Isn't this basically what Luminar does?
Click to expand...
I don't use Luminar but I imagine so. My only thought was that the algorithm and manual efforts are likely different approaches to the same end. Coming back to OP's question, I doubt there is a professional market because appropriate tools and work flows already exist, and I don't think what OP is describing adds anything. I just don't think there's much of non-professional market, as suggested by the other responses.
 
deep diver said:
This is a picture of Mrs. Diver and me. It happens that we were at a party for this friend's daughter. We were standing in front of a plain green screen. He uploaded the photo of us to the computer which put in the image of the Rock and Roll Hall of Fame wherever it saw green, and then sent it to the printer. Had I been wearing a light green suit, you would see the Hall of Fame there as well.
Click to expand...

I was thinking of the use case for this API being exactly the same as the use case for the software he was using. As I had mentioned, it already exists, but that doesn't mean OP couldn't provide a similar service for cheaper, or a better service (presumably the only improvements could be in the user experience side, since it already works pretty well).

Of course, if you've already got a green screen there's no reason for it to be offloaded to a server. It's pretty easy processing to just do on-device! Even a hundreds of photos with a green screen shouldn't take more than a few minutes on any modern computer.
 
mackmgg said:
I was thinking of the use case for this API being exactly the same as the use case for the software he was using. As I had mentioned, it already exists, but that doesn't mean OP couldn't provide a similar service for cheaper, or a better service (presumably the only improvements could be in the user experience side, since it already works pretty well).

Of course, if you've already got a green screen there's no reason for it to be offloaded to a server. It's pretty easy processing to just do on-device! Even a hundreds of photos with a green screen shouldn't take more than a few minutes on any modern computer.
Click to expand...
You would still need editing software unless you just want an empty image.
 
mackmgg said:
I was thinking of the use case for this API being exactly the same as the use case for the software he was using. As I had mentioned, it already exists, but that doesn't mean OP couldn't provide a similar service for cheaper, or a better service (presumably the only improvements could be in the user experience side, since it already works pretty well).

Of course, if you've already got a green screen there's no reason for it to be offloaded to a server. It's pretty easy processing to just do on-device! Even a hundreds of photos with a green screen shouldn't take more than a few minutes on any modern computer.
Click to expand...
I understand. I think where I'm stuck now is that some of us are having this conversation and OP appears to have ghosted. We still know almost nothing about his software.
 
You must log in or register to reply here.
Register on MacRumors! This sidebar will go away, and you'll see fewer ads.
Top Bottom
Back