Intersect function

[TVSportsDir]

One of the functions I need in a vector graphics program is the ability to take a photo, put a non-filled shape over the photo, and then using the intersect command, wind up with a cutout of the shape defined area of the photo.

 

I have tried this with Designer and it doesn't seem to work.  I use a standard .jpg photo, then add a rectangle shape over one person's face in the photo.  The rectangle shape over the face is brought to the front in the arrange menu.  I then click to make sure both the photo and the rectangle shape are active and then hit the "Intersect" button.  All I get is a white box.

 

In iDraw, this works perfectly.

[MattP]

Hello! :)

 

Interestingly, there are at least 4 ways you can achieve this effect in Affinity Designer and I think you've picked the only one that I can find an oversight with, sorry! :( I'll sort it out in a minute, but here are the other alternatives that all work (and are useful to know!)

 

Option 1 - Select the image and just use the Vector Crop Tool to modify the image's cropping area.

Option 2 - Open the layers control. Drag the thumbnail that shows your rectangle onto the thumbnail that shows your image and let go. This has cropped your image to the rectangle. You could use this technique to crop to an arbitrary shape, group of shapes, or even another image (with alpha or it will just act like a rectangle!)

Option 3 - Open the layers control. Drag the thumbnail that shows the image onto the text label of the row that shows the rectangle and let go. This has placed your image as a clipped child of the rectangle. It's different to the cropping case described above because your rectangle is still drawn too, so if it had a colour and your image had transparency you'd still be able to see the rectangle's colour through it. It also allows you to set a line style on the rectangle which will show as a neat border around your image. Again, you can add any object as a child of any other object in this way and expect it to clip to its outline which is very powerful.

Option 4 - The method you suggested: Select both objects and perform the intersect command on them. The reason I called the problem here an 'oversight' is that it's doing something consistent here, but unhelpful... Basically, it is producing an object from the intersection area and then setting in the fill from the object originally at the back (the image) - except the image didn't have a fill! We allow you to have a placed image and easily recolour it by just choosing a colour while the image is selected, and it is that fill that we were taking from the image when we created the intersected area. What we actually needed to do was to set in the (potentially recoloured) image from the source as the fill instead.

 

This is obviously very easy to resolve, it's just that I've never done it this way before - I've always done one of the other methods!  :rolleyes:

 

Thanks for the report and expect it to be sorted out soon :)

Matt