Looking forward to that 👍

It shouldn't take you more than 2-3 minutes for every cut once you get the hang of it. Of course, at the start it'll be a bit slower, naturally, but the learning curve is very gentle 🙂

My two new drafts. Any tips on the second guys hair? - his left hand side specifically There was a lot of green pixels on the source image 😢

Both look quite good, but there's a bit of margin for improvement in both 🙂

There's a bit of snow in the first guy's hair, especially on our right. There's a video in the first post on how to easily spot and clean it.

As for the second cut, it looks fine, but I think you may have overdone some clone stamping of the hair at the top and possibly on the right, because it seems a bit lacking in definition. Can we see the source? Also, there's a video in the last spoiler in the first post on how you can remove most of the green reflections in a source if you wanna take a look 😉

Second attempt + the source

Happy to see you've taken the feedback on board and both look much better; more than ready to be uploaded on the submissions forum, frankly 👏

Are they taking you long? Done right, it should only be a couple of minutes' work each image, but they'll take a wee longer initially naturally 😉

@mons  just haven't got the time atm to do these as much as I'd like to contribute. It's just as and when I find a spare bit of time.
I haven't got PS so I'm just learning my ways around photopea. Most “time consuming” stage is sorting out snow or any pixel discolorations such as around the edges of hair etc.
Here is another draft, this was fairly quick although the “hair fixing” stage almost made me gave up on this cutout…
 

Looks fantastic 👏

I don't think it's an issue in this cut but keep in mind that there's a difference between snow and light reflecting on the hair edges. The former should be removed, but not the latter - save yourself some time 😉

So yesterday I just uploaded a new pack which all the sources got rejected. But I do have a website (imageenlarger.com) which I did use for some cuts to make the sources bigger, and it gets the job done. I put some examples below. But can I still work on these rejected sources if the result is good?

@NassFas, if the person doesn't currently have a cut-out in the megapack, then they will be added. However, in my view, all 3 sources are much too small for a 250x250 cut anyway since the final cuts are all not very sharp, especially the last 2.

If the person currently has a cut-out, then they will be compared to the existing cut. If the existing cut is of a good quality, recent and looks similar, then I have to say it is unlikely they will be considered as improvements, in view of the above reasons.

Remember, more recent is nice, but it is not always automatically an improvement if the source or cut quality isn't there.

what platforms or fonts do you usually use to get photos of players?

Generally I look on the teams website, facebook or google

There's a whole thread on the best ways to find sources for your requests 😉

Do you have a specific topic?

There's a link in my previous post, do you not see it?

How is this @mons ? Wanted to ask before uploading. I'm not sure if the collar is needed (I did watch the video).

Perfectly fine and almost exactly what we look for in our cuts 👏

One small thing you might want to keep in mind for future cuts is to rotate the image before cropping it, in the manner demonstrated in the below video. It's barely perceptible here but the face might be a bit more tilted in other cuts you do, so I thought it prudent to let you know about it 🙂

Perfect! 

Thanks for the heads-up. Will definitely look out for that.

Would you mind uploading it for me @mons ?

Did not know that there was a limit of requests (will of course remember next time I request a lot of faces).

Edit: I found a way to make it work. Just have to do some more cuts for it to work and upload as a pack.

Better regarding the collar? @mons 

Yep 👏

No need to bring it that much upwards, showing a few pixels is enough 🙂

Alright. First time doing it, so will take that into consideration next time. 

How come many parts of the hair got cut off? I cropped the image (it was very large though)

@NassFas, if you had any idea how hard it was and how long it took to do a cut from an image with hair like that prior to remove.bg, I guarantee you that you would not even think to ask that question 😀

The image at the top is a perfectly fine, I would even say excellent, cut-out of the source image. The few strands of hair missing at the top are insignificant, wouldn't you agree?

Anyone know how I can fix this? I used remove bg to crop the image. Is this just a problem with the source and I have to find another picture? 

@Ahmi, if you're using a source which has already had the background removed, in some instances this might not be avoidable when using remove.bg. What you could do is to use the methods displayed in the tutorial on snow (which is what we call the effect in the first image) from the first post. For images like this, putting it through remove.bg a couple of times has also been known to remove a good proportion of it.

I know this is just a draft cut, but when you're completing it, try and crop it even closer to the chin or facial hair. We aim to have no more than 2-3 pixels beneath the chin and I suspect that having it like it is in the second image would be a bit more than that once it is downsized to 250x250. The reason we ask that is so the cut-out takes up as much space of the canvas as possible. After all, it's a facepack not a face-and-neck-pack 😉

I think I was able to fix the issue. Also I moved the collars up a little bit more. Do you think this is good enough for the final product? 

Ideally, there would be a bit more hair details, but if the source was pre-cut, you can't just magic them up out of thin air. Looks perfectly fine otherwise 👏

Yeah I understand. Unfortunately the photoshoot media for this team appears to be pre-cut so I must make do with what is at disposal, but it's good to hear its fine. I'll submit them in tiny bulks. Also I had a couple questions regarding collars for some of the upcoming pictures I am doing. What are the guidelines in regards for too much collar, for example more wide/long collars, and those that are higher up. Also for source images where the player is at an angle and only one side of the collar is bulky with the other side having no collar. Picture for reference.

For this source, I would recommend rotating the image so that it's upright and not tilted to one side. In this case, probably 6-8 degrees anti-clockwise, I would image. This would probably see the collar being level with his chin, but in cases where it isn't, you'd just nudge it up (as per the video tutorial in the collar spoiler in the first post). Ideally there is a collar on each side, and you can create one yourself using the video tutorial in the last spoiler in the first post 😉

I wanted to show you a program which I use for my newer cuts. It is called hotpot.ai Image Restore. I use it for blurry sources, and for small ones too (first I enlarge with imageenlarger.com) It automatically enhances the colours too, so I no longer have to play with the curves. The result is astonishing 😁

I'm not sure the colour in the first one are better. Brightening and adding a small bit of saturation for the image on the left would be enough, I reckon. The second one looks good though 👍

As with all tools, it's important not to over-use it and to use it only when warranted 😉