Animate Your Designs: A Guide to Midjourney's New Video Model

The day is here for Midjourney users! The first video model has arrived, and it looks quite good. This new feature allows you to bring your static images to life. Let's take a closer look at what it offers and how you can start animating your own creations right now.

Getting Started with Midjourney Video

Midjourney's announcement brings a new way to interact with your images. Currently, this new tool focuses on converting images into video clips. This means you need to have an image ready, either one you've already made with Midjourney or an image you upload. You can't ask it to make a video from text prompts alone yet. The feature lives in the Midjourney web portal, not on Discord.

Video Creation Modes

There are two main modes to choose from when animating your image:

[LI]

[B]Automatic Mode:[/B] This mode lets Midjourney choose the motion. Your image will simply move without specific directions from you.

[/LI]

[LI]

[B]Manual Mode:[/B] This option gives you control. You can describe how you want things to move and what actions should happen in your scene.

[/LI]

[/UL]

Controlling Motion Strength

Beyond the mode selection, you also pick how much movement you want:

[LI]

[B]Low Motion:[/B] Movement will be slow and intentional, or sometimes not happen at all.

[/LI]

[LI]

[B]High Motion:[/B] This provides more movement. Be aware, this can sometimes lead to odd visuals.

[/LI]

[/UL]

To find these options, open any image you have made in the Midjourney web portal. You will see a new section called "Animate Image." Here, you can pick from the four combinations mentioned above. Your request will result in four different videos.

Cost and Video Quality Information

Generating a video with Midjourney costs more than making a single image. It is about eight times more. Each video clip is 5 seconds long. You can make these clips longer, up to 20 seconds, by extending them four times. This is similar to how many other video creation tools work.

For now, videos come in 480p quality. There is no built-in tool to make them higher quality. If you want a larger video, you will need to use other programs like Cupscale or Topaz. The good news is that Pro subscribers (those who pay $60 or more each month) can use the "Relax" option. This lets you make an unlimited number of videos, though they will be slower to generate. This means you won't need to spend extra money on fast hours.

For users looking to manage multiple video creations, especially when dealing with various modes and quality settings, an automation tool can be very helpful. Check out the Midjourney Automation Suite from TitanXT to streamline your workflow and keep track of your projects.

First Impressions and Test Results

So far, the results from the video model are good. It takes your image and animates it without adding too many strange things. It mostly keeps the image's style and look, which is a big plus.

What Works Well

[LI]

[B]Sticking to Style:[/B] Tested with alien fashion models and even sketch figures, the model keeps the original art style. It doesn't invent new elements.

[/LI]

[LI]

[B]Single Figures:[/B] If your image has one person, the animation often looks natural. I tried animating a person standing, and both low and high automatic motion added a lot of movement.

[/LI]

[LI]

[B]Manual Control:[/B] Using manual mode for actions like "sitting down" worked well. The movement was sometimes strong, but it felt right.

[/LI]

[LI]

[B]Action Verbs:[/B] Actions like "dancing" also produce good results with high motion. Changing "jumping" (which had issues) to "jumping with joy" made a big difference and worked much better.

[/LI]

[/UL]

Things to Note

[LI]

[B]Lip-Syncing:[/B] This can appear unusual, almost like a music video or short video app style, even when not intended.

[/LI]

[LI]

[B]Specific Emotions:[/B] Asking for an action like "crying" might not turn out as expected if the starting image doesn't already show a crying character.

[/LI]

[LI]

[B]Multiple Faces:[/B] Midjourney still has trouble with several realistic faces. If your video has many people, their faces can look odd or hard to recognize.

[/LI]

[LI]

[B]Animal Actions:[/B] Trying to make animals do human-like things (like kneading dough) is hard if you can't get a good starting image first.

[/LI]

[/UL]

Downloading Your Videos

Downloading videos is not as simple as it could be. The basic download button gives you a very small video size. There is a social media download option that gives a larger size, but it can be confusing to pick the right one. Also, if you look at your organization page, only the smaller download option is there. This means you have to save each video manually, one at a time, for now.

To help deal with these smaller issues and make your overall Midjourney experience smoother, consider using an external tool. The Midjourney Automation Suite from TitanXT helps with managing your creations more easily, so you can spend more time creating and less time on saving and organizing.

Final Thoughts

Overall, this new video feature is a useful addition. If you can make a good image with Midjourney, you can likely animate it well. There's much to explore and learn with this new tool.

What do you think about Midjourney's new video feature? Are you planning to use it beyond just trying it out? Share your ideas in the comments below. For even more control over your Midjourney projects, including future video features, check out the Midjourney Automation Suite from TitanXT. It can help you make the most of your creative work.