Japanese company Animon released an AI video generator designed for animation, hoping that professional animators will use it. However, can this tool really change the future of animation? Or is it just another hype? Let’s delve into it.
The Current State of AI Animation Tools
Artificial intelligence (AI) has made waves in various industries, and the animation field is no exception. Recently, a Japanese company called Animon introduced an AI video generator specifically designed for anime, sparking questions about the role of AI in animation production. This free tool allows users to input a static image of anime, cartoon, or CG art and create a five-second animation based on a prompt. Animon claims it can help professional and amateur animators speed up the creation process, requiring only one hand-drawn frame per five seconds of video instead of hundreds.
But the question is, does AI have a genuine place in the animation industry?
The Challenges of Animation Production
To fully understand the potential of AI, we first need to understand the challenges of animation production. Even without the involvement of AI, there are already many ways to cut corners in the anime industry. Currently, much of the work is outsourced to Korean animation studios, and the time and money invested are constantly decreasing.
When studios underinvest in animation, it affects anime in two ways:
- Reduced Movement: The movements of animated characters become less fluid, appearing stiff and unnatural.
- Decreased Art Quality: Character designs may look beautiful in promotional posters, but the frame-by-frame drawn versions lack detail or proper proportions, and in some cases, they even look comical.
This decline in quality can disappoint viewers and affect the overall enjoyment of the anime.
Animon.ai: A Deep Dive
Animon.ai aims to help animators save time by reducing the number of frames they need to draw. However, based on the content generated by Animon.ai, it is questionable whether it can truly achieve this goal.
The tool can take an image provided by the user and make it move for five seconds, but that’s almost the only thing that’s guaranteed. To assess Animon.ai’s capabilities, I conducted some experiments.
Experiment 1: Jet Black
I started by giving it a static image of Jet Black from Cowboy Bebop about to drink from a whiskey glass. My prompt was simple: "The character in the image takes a drink from the whiskey glass he is holding."
However, the returned video had several issues. First, Jet Black did not drink the whiskey. He seemed to be talking and bringing the glass closer to his face, but I didn’t see anything that looked like actual drinking. However, the whiskey in the glass seemed very busy - the glass was both filling and emptying, even though there didn’t seem to be any drinking happening.
The AI model also struggled with Jet’s scar. The bottom of it disappeared, and when he opened his eyes, the scar went directly through his eyeball. Because I grabbed the static image from a video on Crunchyroll’s YouTube channel, there was a CR logo in the upper right corner. However, in the AI video, the text seemed to have turned into “Crunchyolo”.
Experiment 2: Maomao
Next, I tried giving it a static image of a dance scene featuring Maomao from The Apothecary Diaries, and I got a very strange result. Unlike the psychedelic, gooey, warping movement from the 2D Jet video, the AI gave me what looked like a 3D model.
Animon.ai is designed to work with CG art, so I suppose it makes sense that it can generate 3D models, but I was very surprised. While The Apothecary Diaries makes heavy use of 3D models in its buildings and backgrounds, the characters are definitely drawn in the shot I used.
Either way, when it came to the 3D model, the generated video seemed noticeably more stable. The movement looked less wobbly, and the model kept a more consistent size and shape of the character. I still wouldn’t want to see it in the show, but it was undeniably a step up from the 2D-style content it gave me.
However, Animon.ai’s sales pitch is all about helping animators save time by drawing fewer frames – which is irrelevant when it comes to 3D animation. It’s already fast and cheap compared to 2D, and it doesn’t make much sense to create a 3D model and then use an AI video generator to animate it haphazardly instead of using animation software.
Limitations of Animon.ai
Animon.ai reveals several limitations of current AI animation technology:
- Unrealistic Movement: The movement generated by AI often looks unnatural and unfluid, which can affect the overall quality of the animation.
- Lack of Control: Users have limited control over the outcome of AI-generated videos, making it difficult to achieve specific animation effects.
- Inconsistency: AI models may produce inconsistent results, leading to unexpected glitches and errors in the animation.
- Limited Applicability to 3D Animation: The value of Animon.ai for 3D animation is questionable, as there are already more efficient ways to produce 3D animation.
The Future of AI Animation
Despite the limitations of Animon.ai, AI still has potential in the animation field. As AI technology continues to develop, we can expect to see AI tools become more sophisticated and powerful. In the future, AI may be used to automate certain aspects of animation production, such as:
- In-Betweening: AI can be used to generate in-between frames between keyframes, reducing the number of frames that animators need to create.
- Background Generation: AI can be used to generate detailed and realistic backgrounds, saving animators time and effort.
- Facial Animation: AI can be used to automatically create realistic facial animations based on speech.
However, it is important to remember that AI should be seen as a tool to enhance human creativity, not to replace it. The most successful applications of AI animation are likely to be in scenarios where human animators collaborate with AI tools to create high-quality, engaging animations.
Limitations of AI Video Generators
Overall, I think AI video generators aren’t ready for any practical work for two main reasons. One is that the motion looks too bad, and the other is that you have too little control over the results. This is the same as any other generative AI tool – they aren’t consistent and responsive enough to be reliable. We might get there someday, and it’s still fun to try out early versions of the tech, but don’t let the salespeople fool you – these tools aren’t ready for commercial use yet.
Final Verdict
If you know nothing about animation and you want a quick GIF or a video for personal use, this tool might work OK (or if it doesn’t, it’ll at least give you a good laugh). However, it’s hard to imagine any amateur or professional animator seeing real value in this.
If you want to see what the tool is capable of, the Animon YouTube channel released a music video that looks almost 100% AI-generated – or at least, it looks like it.
Conclusion
In conclusion, Animon.ai may not be the future of anime. While it may offer some limited benefits, such as generating quick GIFs, it is far from being able to replace human animators. The limitations, such as poor movement quality, lack of control, and limited applicability to 3D animation, make it an unviable tool for the professional animation industry.
However, the field of AI animation is still evolving. As technology advances, we can expect to see AI tools become more sophisticated and powerful. In the future, AI may be used to automate certain aspects of animation production, such as in-betweening, background generation, and facial animation.
Ultimately, the role of AI in animation will depend on the ability of human animators to collaborate with AI tools. By leveraging the strengths of AI while maintaining the core of human creativity, we can create animations that are more engaging and visually stunning than ever before.