PS adds text generation AI function! Becoming a PS master only needs these three steps...

How many steps does it take to become a PS master? Answer: Three steps.

First step : Lay the foundation and master some basic knowledge of PS;

Second step : To improve the ability, to be familiar with the PS toolbar and various functions through practical operations;

The last step, It is also the most critical step, that is two words - "show off skills".

Master shortcut keys The use of the keyboard, regardless of efficiency, the keyboard is crackling, which seems to be operated by a "master".

Of course, need There are a little more shortcut keys...

The above is what becomes The trilogy of PS Great God. How about it? Is not it simple".

Want to be" "PS Master" is of course not that simple, from getting started to giving up is the normal state of PS learning.

But when PS When encountering AI, ordinary people's "Road to PS Advancement" has a new turning point. To become a PS master, you only need these three steps.

First step , download and install PS;

Second step , import pictures;

The third step is to hand it over to AI.

Following the PR ( After Premiere Pro), Photoshop also introduced new AI tools.

May 23 Today, Adobe released the latest version of Photoshop24.5, the most attractive of which is the AI ​​tool called "Generative Fill" (generate fill). This technology is supported by Adobe Firefly and works similar to the "inpainting" technology in DALL-E and Stable Diffusion.

User selected After specifying the selection area, input your own thoughts into the text box through text, and the system will perform corresponding operations through the analysis of natural language. In the past, operations such as seamless erasing, background changing, denoising, matting, and presets, which were relatively complex and depended on proficiency, can now be completed immediately by entering a few words.

Ordinary people want To become a "PS God", is really just "talking about it".

But Adobe said ,Generative Fill is currently in beta,This feature will be officially added to Photoshop later this year. In addition,Generative Fill currently only supports English language, Chinese language is not applicable.

01

From bottoming to finishing:

Perfect solution to the "edge problem"

To be God of PS, Wuhen P-picture is a very deep knowledge.

However, from Judging from the official demonstration video, Generative Fill solves this problem almost perfectly.

Added Generative Fill mainly includes three major functions: adding new content to the image, deleting/replacing part of the content of the image, and extending and drawing the existing image outward.

Official release In the promotional video of , by selecting a part of the picture, and then input the content you want to add "yellow road lines" (yellow road lines), click generate (generate), and the yellow road lines will appear in the picture.

And the picture Zoom out, select the blank space on the edge of the layer, click Generative Fill, and then the system will automatically fill in the corresponding content according to the tone and content of the picture.

In addition, the The picture also has the AI ​​automatic recognition function. You only need to click select subject to easily select the target object in the picture, instead of using the lasso tool to pick a little bit as in the past. Even if you are very careful, the edges of the picture It is still difficult to escape the "dog gnawing" ending.

If you want to replace Background, then deselect the background, then enter the desired scene, click generate, and the subject and the new background can be perfectly blended.

and new features It also supports adding new images on top of the original image.

But from the official It can be found in the demonstration that the "new content" it adds is not "new", but is actually based on the "old content" already available on the Internet. AI tools need the support of the material library, and the richness of content elements depends on the size of the material library. If there are few resources in the material library, it is very likely that the found picture does not match the style and tone of the original picture.

While beta The material library is still relatively limited, but it will definitely be continuously improved before the official version is released.

From the current From the perspective of the version,The newly added AI tool is on the "edge processing problem" of the previous P map It is really eye-catching. The cumbersome process of seamless photoshop has been greatly simplified. What used to be a few hours of work now only takes a few seconds, which makes PS creation more friendly and close to the people.

02

Adobe's "killer feature":

image copyright

Talking about AI The material library of the tool will inevitably involve another important issue-The copyright of the picture< /span>.

No publicity Where does the image library I use to train AI come from? It may contain a large number of unauthorized images. This is a common problem with AIGC tools currently on the market. This may not have much impact on ordinary users, but in commercial use scenarios, copyright The problem has buried a lot of hidden dangers of infringement for users.

To this, Adobe claims thatonly used content that the company has the right to use for training—such as Adobe Stock images, open license content, and content without any copyright restrictions. In theory, this means that anything generated using the Generative Fill feature can be safely used as business use.

At the same time , Generative Fill also supportsContent Credentials< span style="color: #555555; --tt-darkmode-color: #555555;">, a "nutrition labeling" system that attaches attribution data to images before they are shared online to inform viewers< strong>Whether this content was created or edited using AI.

AI technology The use of security is an important factor for AI companies to widen the gap. Although Generative Fill has little difference in the function of AI tools such as Midjourney and Stable Diffusion, and it was released later in time, it relies on security advantages and Adobe's huge software. The audience base, Generative Fill has the possibility of catching up from behind.

03

Art creation empowered by AI:

encourages, not replaces

Generative Fill With such a powerful function, it will inevitably cause people to worry. Are PS designers going to face unemployment? Does this stifle artistic creation?

In fact, The application of AI tools is mainly for creative efficiency and will not damage artistic creativity. On the contrary, its It also stimulates artistic creativity.

The power of Photoshop The function is recognized, but the high technical threshold makes it an "aristocratic art", which is limited to a few people.

Generative Fill The application lowers the technical threshold of PS and saves a lot of preparatory work and adaptation process. No matter what experience and skill level content creators are, they can generate ideal content according to their own language, 3D stereoscopic effect, color gradient and Complex techniques such as light and shadow matching can now be easily applied.

Adobe will generate The powerful combination of the speed and ease of use of AI and the powerful functions of Photoshop brings about the liberation of artistic creation,People's artistic imagination is no longer limited by complex and cumbersome techniques, and more energy can be concentrated on artistic creativity, thus making artistic creation more free, efficient and interesting.

Adobe Digital Media "This will have a transformative impact on content creation and monetization of content creation," said President David Wadhwani.

Author: Watermelon Typesetting : Luo Weiling

The picture comes from Q Aberdeen earned it by surfing the Internet, if there is any infringement, please contact the background, and Q Aberdeen will delete it~