Up next in 10
Canva Text to Image_ The ULTIMATE Prompt Guide
571 views
Jan 14, 2024
This video serves as the ultimate guide to using Canva's Text to Image feature, providing comprehensive instructions on how to effectively generate images from textual prompts. The content likely covers various prompts and demonstrates the capabilities of Canva in transforming text into visually engaging images. The video aims to assist users in maximizing the potential of Canva's Text to Image feature through a detailed and informative guide.
View Video Transcript
0:00
Who can tell me what the difference is between this photo of a young boy with glasses and this one
0:06
Or even this one. The answer is simple. The prompt. I spent a couple of days researching how to get the most out of text to image, Canvas AI Image Generator
0:17
And I can say for sure that what I discovered helped me get way better pictures
0:22
So yeah, I think I'll crack the code. And in this video, I'm going to split it all out for you and teach you everything I learned about
0:30
about how to prompt Text to Image. Let's dive in. All right, guys, let's start at the beginning
0:36
and see what is Text to Image. So Text to Image is Canvas version of Mid Journey or DalaE
0:46
but it's baked in the Canva experience. Now, when you look at this from inside
0:52
the editor in order to find it, you have to locate the little apps tab
0:56
right here on your object panel, click on it, and it should also, be the first one that you see here at this feature. So if you click here, you can start prompting
1:04
by typing a description in this box right here. You can see some examples. And yeah, once you
1:10
have a prompt, you can select one of the styles. So if you click on C, all you see you have
1:15
photography, digital art, and fine art. So different kinds of stars are predefined for you. And the last
1:22
thing you can do is select the aspect ratio right here. So you could have square, landscape
1:27
or portrait. So that is in a In a nutshell, what text to image is
1:32
One thing I realized while doing my research about text to image is that if I want to understand how to best write my prompts, I need to know what's under the hood, meaning what is the underlying technology Canva relies on for text to image
1:48
What powers it? Because Canva is not an AI company. Canva is a design company
1:53
So they probably rely on somebody else's API, aka technology, to bring
1:59
us this text to image app. So I did a little bit of digging around and easily discovered that the
2:05
technology behind text to image is actually stable diffusion, which is one of the most famous
2:11
AI image generator model by a company called Stability AI. So what Canva did is that they made a deal
2:18
with Stability AI in order to be able to use stable diffusion inside Canva. And for safety reasons
2:25
Canva added a couple of security layers or filters on top of the base
2:29
model so that people cannot prompt for things that are dangerous or obscene or you know
2:35
inappropriate and finally to improve the overall experience canva came up with the predefined
2:41
styles so that you don't have to specify that you want a photo or concept art or a painting you can
2:46
just click on one of their predefined styles and avoid the hassle of including that into your text
2:53
prompt all right we are going to get into the crunchy part and i'll show you what i learned about
2:59
better prompting in a second. But before we get started, I believe there are a couple of things you should know
3:05
And the first one is that text to image will work way better when you prompt in English
3:10
Now, English might not be your native language, but try to prompt in English for the best result
3:15
The second thing is that text to image is available to all users
3:19
so pro and free, but, and there is a but, you will be limited to 100 images per day
3:25
And 100 images is about 25,000. prompt because for each prompt you will get four results. So four times 25 is 100 images generated
3:36
with text to image. So be mindful of that because that limit is easily reached. The third thing
3:42
you should know is that it works better when you talk to it as a human versus talking to it as a
3:49
computer program. Okay. So don't try to create two elaborated prompts. Sometimes simple short prompts will
3:56
work very well. So just talk to it as you. would to a friend. And then the fourth thing, which is probably the most controversial of all
4:03
is the copyright issue. Can I use stuff that I generated with text or image for commercial use
4:09
I just happened to talk to a friend of mine who is a copyright lawyer yesterday and she didn't know
4:14
As of today, experts don't know what to do with generative AI in terms of intellectual property
4:21
So if you want to be safe, just don't use them for commercial use right now and wait for the dust to settle
4:26
and the legislators to come up with some sort of regulations. What you really shouldn't do is ask me that question in the comments because I don't know
4:35
All right, now that we get the formalities out of the way, let's see how to best prompt text to image
4:44
So how are we going to do this? Well, based on my research, I came up with two simple steps to really improve the quality of your prompts
4:52
and therefore the quality of your outcomes with text to image. The first step that we are going to explore right now is a specific prompt library that I discovered with a bunch of stable diffusion ready to use prompt
5:06
So we are going to dive into this right now. And then the second step we are going to explore a bit later in the video is a formula that has been working very well for me and my prompts
5:15
But let's start with the prompt library I was telling you about
5:21
The website is called Lexica and it is a stable diffusion. search engine. All right. So here you will find a bunch of different images that have been
5:32
generated with stable diffusion and the beauty of it, for example, if you like this rabbit on a
5:37
motorbike right here, the beauty of it, if when you click here, you should see the entire prompt
5:42
that the person has been using to generate that image. So from here, you can easily copy that prompt
5:48
with this button right here. So you can do that, then come back to canvas text to image and paste
5:53
that prompt. You will not get exactly the same image because this process is pretty much randomized
5:59
but it gives us a good idea of what people are using in their prompts to generate specific images
6:04
So if you like something like this, for example, you have the entire prompt right here
6:09
Okay, so let's try one. Let's copy this prompt. Okay, so this is a lost astronaut and Jupiter epic
6:16
scene displaying in the background the moon and other planets and stars and Killian Eng
6:22
vibrant colors, dynamic lighting, blah, blah, blah. There's a bunch of things
6:27
Then there is some extra information that I thought I would cover
6:31
It says trending on Art Station. So ArtStation is another website where people like to show their digital images and creations
6:39
And then it says art by Zislav Baskinski. Probably pronounced that wrong
6:45
So that defines the style of the picture. Okay. So this guy is an artist who creates digital art
6:52
So by giving the AI the name of a specific artist, it gives it direction
6:57
It gives it a specific style to emulate. All right. So I copied this style and now I want to come back to Canva and copy that prompt right here in my text to image prompt box
7:09
I'm not going to change any of the styles for now. I'm just going to go for the square format and create my image to see if this comes even close to what I saw on Lexica
7:21
So these are the results that have been generated by text to image
7:25
Let's click on a couple of them. So this one right here, kind of like the same mood
7:31
And let's use this other one right here. So these two have been generated by text to image
7:37
I would say they are not as good as what I saw on Lexica, but they're not bad either
7:44
Let me upload the one from Lexica right here There you go So yeah if I compare The original one this one with these two other ones they are pretty similar
7:57
So browsing through Lexica and the thousands of images that you can find there to find an image that you like and then simply copy the prompt and paste that into canvas text to image is a very simple way to generate stuff that you know will look good because it already looks good on Lexica
8:15
Okay. a couple of more things I want to show you on lexica and I'm going to try another prompt to show you
8:20
kind of the variety of what you can do. If you look under the image, the main image that is
8:25
selected right here, you should see different variants of the same image. So these are the other
8:32
images that have been generated by stable diffusion. The first one being usually the best one
8:38
the one that the person who prompted selected out of the four options. So usually the first one
8:42
will always be the best one. Something else is that you will also see a negative prompt most of the time
8:48
Just don't worry about this one too much because canvas text to image doesn't have a field
8:54
for negative prompt, while stable diffusion does have that field. All right, let's close this one and come back up here because I want to show you a couple more
9:02
things about this stable diffusion search engine, because that really is what lexica is all about
9:09
You have filters right here. And in the filter, you can decide to choose either lexica aperture or stable diffusion 1.5
9:18
So both lexica aperture and stable diffusion 1.5 are two different image generation models
9:25
Lexica aperture being the newest one. So it should render the best results
9:29
And I'm not completely sure which of these two models canvas text to images relying on
9:36
So I couldn't find that anywhere. If you know, let me know in the comment section
9:40
But for now, what I want to show is. you is that when you click on that filter button, lexica aperture is the default button and
9:46
will give you some better result in my opinion. Let me just show you what you get from the
9:52
stable diffusion 1.5 model. You see the photos are not exactly as good. So I would definitely
9:59
recommend you use lexica aperture in terms of browsing lexica so that you can find the best
10:06
possible images. All right, let's find another example before we move on. Let's find
10:11
a cute photo maybe something more in the style of a photo I can also search right
10:16
here so let's say portrait of a young boy smiling let's see what we can find
10:26
okay a bunch of different portraits of a young boy smiling yeah let's try this one
10:32
right here it says gotham adani a kid sad old color oil painting realistic okay
10:38
this seems like a good prompt I'm going to copy this prompt Come back to Canva. Add a page here. Come to my text to image and simply going to paste that prompt
10:48
Gotham Adani. I don't know Gotham Adani. So I might just look it up quickly
10:52
Gotham Adani is an Indian billionaire industrialist who is the founder and chairman of the Adani group
10:58
Alright, so it's just some rich Indian dude. Okay, so we have the original portrait here
11:03
Let's see if text to image can generate a similar portrait of Gotham create
11:08
And you will notice that I didn't use any of the filters. I just want to see how it works when I prompt kind of like without any other information
11:17
So I do have some cool portraits here. I really like this one
11:21
This one is nice. And I really like this one as well
11:25
So these two are pretty cool. Let me grab the original one to make a side by side comparison
11:32
So there you go. It is uploaded. And yeah, we do have something pretty similar in terms of the texture, in terms of the style
11:41
oil painting, text to image recognized that this prompt worked very well
11:46
So oil paint, not even oil painting, it's just oil paint. And I get a pretty similar result
11:53
All right, I don't have the bindi right there for Gotham, but it does look pretty similar
11:58
in terms of the kind of result that I achieved. Okay, so you see this technique works very well
12:03
Before I jump to the second way to improve your prompts, I want to show you a couple more
12:07
images I was able to generate following the exact same strategies. So these two ones right here about Mark Zuckerberg, I found based on this other photo
12:17
from Lexica. So I grabbed the prompt, came back to text to image, put it there, and these are the
12:23
two images that it generated. So again, the original photo and the two ones that were generated
12:29
Similarly, I found these vintage photo portrait of Elon Musk on Lexica with this prompt right here
12:35
So I copied that prompt, pasted it into text to image, and these are the two images that it generated
12:41
pretty mind-blowing in my opinion. Like, Elon Musk is really recognizable on these photos
12:46
And yeah, when we look at the original one, this one right here, they're pretty similar
12:51
So again, I was very satisfied with this result. And you know me, I'm a big sneakerhead
12:57
so I also had to try generated high-resolution images of sneakers and specific models of sneakers
13:04
And these are the two images that Canva was able to generate
13:07
for me of the Air Max 1 model. So this one and this one right here
13:12
They're not perfect, but it's pretty darn close. All right. So this is the first way you can really improve your prompt in text to image
13:20
It's simply to find inspiration on lexica. Now, let's move on to the second way you can generate better images with text to image
13:31
In this second way of generating better prompts and therefore better outcomes for your text to image
13:39
We are going to lean on AI copy text generator, such as chat GPT or even Canvas Magic
13:47
Write, and ask them to generate our stable diffusion prompts for us
13:52
So this is just a smart way to use AI to create a better way to communicate with another AI tool
13:59
And this idea, by all means, is not something I invented. I need to give credit where credit is due
14:06
and I learned that trick from my friend Matt Wolfe. Matt's also a YouTuber
14:11
His channel is called Matt Wolfe and he talks about AI and all the latest AI tools
14:17
His channel has been exploding recently because everybody wants to know about AI
14:21
He also has a cool website called Future Tools where he organizes and reviews all of these AI tools
14:29
So I recently had the chance to interview Matt for an episode of Pro Tips
14:33
which is not yet published on this channel So if this is something you are interested in watching, make sure you subscribe to the channel so you don't miss that video when it will be published in a couple of days
14:46
All right. So back to the tutorial. Matt is actually the person I learned this tip of using chat GPT to generate stable diffusion prompts
14:55
And yeah, I think it was in this video right here. It says epic chat GPT prompts, the playbook
15:00
So yeah, if you are interested, I will have a link in the description to this video of Matt
15:06
But yeah, so I learned that trick from him. What we are going to do, we are not going to be using chat GPT to do the same trick, but we are going to stay within the Canva ecosystem and use Canva's magic right, because it works just as well
15:19
All right. So first, let me introduce you to the formula that I learned from Matt
15:25
So the formula goes like this. Here's a formula for a stable diffusion prompt
15:29
So an image of adjective, subject, doing action, all of that between brackets, creative light
15:36
lighting style So that gives us some info about the lighting style Detailed realistic HD trending on Art Station Remember what we saw on Lexica that tag that was there in the style of famous artist one famous artist two and famous artist three
15:52
So what we need to know, everything that is here in between these brackets
15:56
will be kind of like randomly generated by the AI. Okay, so they will know what's in between brackets
16:04
You can have fun with this and play with different alternatives, different options
16:08
So this first part of the prompt is what Matt is using to kind of train chat GPT
16:15
And then he will ask something like, do you understand the formula? Chat GPT will say, yes, I understand the formula
16:20
And then he would come up with the second part of the prompt, which is the actual prompt
16:24
And he will say, okay, then write five stable diffusion prompts using the given formula
16:29
with the subject being XYZ. So XYZ being the subject you want to generate the photo or the image about
16:37
All right, so let's try this with different kinds of subjects using MagicRite to see what kind of prompt it can come up with
16:45
And then we will use these prompts and run that in Canvas Text-to-Image
16:49
All right, I am here in my test document. Let's open the Canva Assistant
16:53
Let's find MagicRite right here. MagicRite is a pro feature, but you will have the chance to try 25 times for free
17:02
But if you have exhausted all of these free tries, you can, one, support
17:07
subscribe to Canva Pro or to use ChatGPT to do the same thing
17:11
It will work just the same. But here, since I want to show you guys how to do everything from the Canva ecosystem
17:17
I'm going to use Magic Right. Okay, Magic Right. So this is my prompt
17:22
Here is a formula for a stable diffusion image prompt. Okay? I changed the formula slightly to kind of adjust it to what I want
17:31
Here, I want a photo as close to a real photo as possible
17:35
So this is how I changed it. photo realistic image of and then I let the adjective subject doing action creative lighting all of
17:44
this I let that as such in the style of and then instead of artist one artist two artist three
17:52
what I did is I did a quick Google search for famous portrait photographers and I grabbed these
17:59
names of famous portrait photographers and I paste them here right there as the inspirational style
18:06
for this outcome. Okay. Write five stable diffusion prompts using the given formula
18:11
with the subject being a young African-American girl. So this is my prompt
18:17
You will see I let some of the random stuff going on, but in terms of the artistic direction
18:23
I gave it the names of famous portrait photographers. All right. So let's see how MagicRite can handle this formula
18:31
and generate my prompt. So here are my five prompts. me make that bigger for you guys
18:39
I'm going to bump it up to 25. Okay, so this should be now readable by everyone
18:46
A photorealistic image of a serene, young African-American girl gazing into the distance
18:52
lit by a dramatic sunset, detailed, realistic HD trending on art station in style of Richard
18:59
Avedan, Yusuf Karsh and Helmut Newton. Okay, so these are the three photographers
19:04
All right, so what is going to change here? is pretty much the subject and the action and the style of lighting
19:11
Okay, so I'm going to highlight this for you and we can choose together which one we prefer
19:16
Okay, a serene young African-American girl gazing into the distance. Then we have a confident young African-American girl dancing in the rain
19:26
Okay, then we have a playful young African-American girl blowing bubbles lit by a whimsical, colorful, blah, blah, blah
19:34
Okay, so blowing bubbles. thoughtful young African-American girl reading a book. Okay, so let's highlight this. And then a fierce
19:43
young African-American girl boxing. So these are my five options right here. I think I'm going to go
19:49
with the first one and then we'll see where we go from here. So I'm going to copy this entire
19:56
prompt, bring it to the prompt box in text to image right here. I'm going to leave it on none
20:01
in terms of the filters. Come to my white page and create this photo
20:06
see what happens. All right. I have some high quality stuff right here. Let's click on the first one right here. This is really really nice. Like this photo is super realistic indeed. I like how she's indeed gazing into the distance in a thoughtful way. This one is also cool. Yep. Like everything is well proportionate on this photo. Like there's no major like bug or extra ear or second nose or anything like that. This is good
20:36
This one also good, though the subject is much smaller. So my preferred one is really this one right here
20:44
So yeah, there you go guys. We see that the outcome from my prompt is of really high quality because I gave it some inspiration
20:53
All right, so before I move on to the next prompt, let me show you what text to image generated for the exact same prompt yesterday when I was prepping for this tutorial
21:02
I was able to generate this one image right there for the exact same
21:06
same prompt. So yeah, the outcomes were pretty mind-blowing in my opinion. Let me show you a couple of
21:12
other photos that I was able to generate using similar kind of prompts just changing slightly
21:18
The subject I was using or maybe the name of the famous photographers. I was able to generate
21:23
this Indian older man wearing a turban and I really like this one. I generated this close-up
21:30
of the face of an Asian woman smiling using the same technique. And this one right here was generated by
21:36
using the same prompt of this older Indian guy wearing a turban but this time I changed the
21:42
style being none to vibrant so when I use the vibrant style this is the image that it generated for me
21:50
all right now that you know how the formula works now that you know how to generate your prompts
21:55
with magic right let me show you how to go one step further and really find the right references
22:02
that will give text to image the right direction when it generates your images
22:07
And the way we are going to do this is that we are going to continue using the AI
22:11
to generate lists of different artists famous in different art categories. The first one being photography
22:19
So let's use Canvas Magic Right again, or it could be Chat GPT
22:22
to generate a list of famous photographers who have won prizes. Okay, so I'm going to copy this prompt right here
22:30
Open canvas magic right. Okay, there you go. And simply prompt, generate a list of famous photographers who have won prizes
22:39
Give me their names only in a bullet list. Okay, so Canva should do that
22:48
There you go. Generated my list. Hansel Adams, Diane Arbus, Henry Cartier-Bresson, Robert Kappa, etc., etc., etc
22:55
So I could take one randomly, Steve McCurry. So I'm just quickly double-checked with
23:00
Google who is Steve McCurry. All right. There's Steve McCurry, American photographer
23:06
All right. So maybe we can see some of his art. And there you go
23:10
We have here a bunch of different photos from Steve McCurry. Okay, I see now the famous photo of this girl right here
23:18
This one has been on the cover of so many different magazines
23:22
So this shows me that the info is legit. I have now a list of 10 famous photographer
23:28
And you can go even more granular. You can ask for portrait photographer You can ask for children photographer You can ask for abstract photographer You can ask for photographer who were published in specific magazines So I have come up
23:42
with this list right here of portrait photographers. I will have a link to a document with all of the
23:49
fruits of that research right here that I'm showing you on screen in the description of the video
23:55
That would be a template link to this canva presentation. So you can use that link in the description
23:59
and have access to all of these resources. So what I did, I did that not only for photographers
24:05
but I also did it for famous graffiti artists who have been mentioned in mainstream media
24:10
And that is what Magic Wright came up with. Banksy, Keith Herring, John Michel Basquea
24:15
Shepherd Fairy, Invader, Ask Jemios, etc, etc. All right, so let's try to generate a prompt this time
24:22
with this list of graffiti artists. And I'm going to be using ChatGPT to just show you the workflow
24:27
using a different text generator. So I pasted my formula. Here's a formula for a stable diffusion image prompt
24:34
What I change is this. I want a graffiti of. Okay, so instead of an image, I say a graffiti of
24:41
And then just let everything random. And yes, in the style of..
24:46
Instead of having three different artists, I only gave it one. So it should be more precise and more faithful to the style of this specific artist
24:55
And the artist I chose is Keith Harris. Okay, so do you understand the formula
24:59
I'm going to run the prompt. I should say yes. Okay, so now I will paste the second part of the prompt
25:05
Write me a five stable diffusion prompts using the given formula with the subject being and then I'm going to change the subject
25:13
I'm going to write children holding hand. Children holding hand. Okay, let's run this prompt and see what's going on
25:23
So a graffiti of joyful children holding hand surrounded by a warm
25:27
golden lighting with a whimsical and playful style featuring intricate details and realistic texture
25:33
and HD quality trending on art station in the style of Keith Herring. So yep, children holding hands
25:39
under soft glowing light. Okay, so the lighting style is going to be different. I'm going to choose one
25:44
randomly. Let's take this one here. Mysterious and atmospheric lighting with mystical and fantastic
25:50
style. Maybe not this one because Keith Herring is not about all of that. Dramatic and moody lighting
25:55
Siminematic. Let's take the first one right here. Warm golden lighting. Okay, so I'm going to copy that
26:01
Come back to my text to image. I'm going to paste that here. All right. So all of the prompt right here
26:09
I'm going to leave it on none for now and generate it to see what it comes up with. Is it going to look
26:14
like Keith Herring? Yes. The first one I see here is yeah, pretty much Keith Herring. So yeah
26:21
I like that. So the graffiti generated were these ones. So we have this
26:28
We have this. That is pretty Keith Herring as well. In my opinion, we have this
26:34
So all of these are pretty good actually and pretty faithful to my original prompt
26:40
So yeah, that was just to show you it works with different styles
26:44
Let me see if I can even refine this further by instead of leaving the filter on none
26:50
if I post it as a concept art. What would happen if I use that
26:55
filter instead of no filter and compare the new outcomes with the previous one
27:01
There we go. So we have the fourth one. My favorite of all of these, probably this one right here
27:08
Yeah. This one is really cool. We can see the texture of the wall
27:12
Let me move these out of the way. And yeah, so this one we can see the texture of the wall
27:17
We can make it bigger. To really see, we see a door. So we can see this is a graffiti
27:23
Let me delete these to have this full screen. This is actually very good because we can see it's the wall of a house and this has been painting
27:32
So if we make this full screen, we could add more elements to this
27:35
We could have people walking in front of that wall, etc., etc
27:39
And talking about graffiti, I wanted to show you two other artwork
27:43
I was able to generate using similar formulas. So this one right here of a city skyline and this other one asking for a colorful city skyline at sunset
27:53
So these two were generated. I was really happy about them. So, yeah, showed you how to generate a list of graffiti artists in the document that I will leave in the description for you
28:04
I will also have some caricature illustrators. So all of these guys right here
28:08
I also left the prompt that I used in Canvas Magic Right
28:12
So you can kind of get inspired by this prompt and generate your very own prompt about some other style of art or other category of artists
28:21
What else did I play around with? Yes, some famous 3D generate
28:25
software because I figured not only the artist will influence the outcome, but also the tools
28:33
used to generate these outcomes. Okay. So in terms of the software, it generated this kind of list
28:39
Autodesk 3DS Max, Blender, Cinema 4D, ketchup, Maya, et cetera, et cetera
28:44
What else did I ask? I ask it to generate a list of famous NFT collections
28:48
And there you go. Five NFT collections that were very famous. The Cryptopunks, the board app, the art blocks
28:55
Puggy Penguins and the world of women. And so I use my formula in Canvas Magic Right and ask it to generate five stable diffusion prompts of a penguin eating a cookie in the style of the Pungy Penguins, the famous NFT collections
29:09
And these are the two best ones that it came up with
29:13
So this was generated by Canvas Magic Right, just using the right prompt
29:18
So I think I'm going to leave it here for you guys
29:21
I think you get the point. You need to, one, use the formula. to ask the AI tool of your choice to generate your prompts for you
29:30
And three, do your research about famous artists, famous art categories, famous techniques
29:35
famous softwares used to pretty much achieve these type of art forms
29:41
And then if you mix out of that, you will get the absolute best result out of canvas text to image
29:47
Right, let's wrap this up with three important takeaways. Number one, the model, the image generation model
29:53
the AI model, on photos taken from the internet, this dataset. So feedback these references to the model to
30:01
generate the best outcome. All right. So that's exactly what I was doing when I was researching
30:08
famous photographers, famous painters, famous graffiti artist. Because it has been trained on
30:15
images with tags like this is Banksy, this is Van Gogh, this is Picasso. It knows that. And therefore
30:21
if you use these keywords, if you use these references, and feed them back to the model, it will generate the best outcomes for you
30:29
Number two, leverage MagicRite and Chat GPT to write your prompts for you
30:35
Because who is more capable of talking to a robot than another robot
30:39
So yeah, ask ChatGPT or Canvas MagicRite or any other AI copy generator to write the prompt for you using the formula I gave you
30:48
Again, credit goes to Matt. And third, grab our prompt guide that will be listed in the description
30:55
This will be a template link to this deck I just showed you
30:58
So by using that, you will have all of the references. Feel free to build upon that, maybe share it in our Facebook community
31:05
so that everybody can benefit from your research. This was my research
31:09
I hope you enjoyed it. Thank you for watching until the end. You are an absolute legend
31:14
If you're still with me at this time, I'm going to leave you with our playlist of the other video and tutorials I created about the other AI tools
31:22
Yeah, it's right there and it's full of gems like this video. So click there
#Clip Art & Animated GIFs
#Design
#Multimedia Software
#Photographic & Digital Arts