Tag

AI

Browsing

This story is adapted from the 8-book novel series, The Computational Universe, Book 1: Xinglong Awakening, with parts of the story drawn from Chapter 1.

It is about Liang Feng and her nightmare, which is also the nightmare of Xinglong. This is an experimental piece, where I experimented with the text-to-video feature of Veo3. It was my first experience with Veo3, and it exceeded my expectations. Because it was almost pure text-to-video, Liang Feng’s face changed a bit between frames. I will also change Liang Feng’s face when I proceed with the series, using image-to-video as my standard process.

About Me

I am Si Liang, an AI filmmaker based in Guangzhou, China. I attended New York University, where I studied philosophy. I then went on to become a programmer for over 10 years and an author with more than 10 years of experience. I am currently working on an octology sci-fi novel series called The Computational Universe.

Subscribe

As I continue to explore my unique tone and storytelling style in AI filmmaking, you may notice that my style evolves throughout this 25-short-film journey. Stay tuned and subscribe to my channel; many more AI short films are on the way.

This is a question raised by a group member that resonates with me. I share similar doubts because current AI-generated realistic styles suffer from a strong uncanny valley effect, with Veo3 perhaps being an exception. In other words, audiences may immediately recognize AI-generated realistic works as artificial, breaking their immersion. They tend to be much more tolerant of animated works.

In the year 2700 of the Xinglong Era, after the victory in the Fourth Stellar War, mathematician Yu Baiyun was deeply concerned about the future of civilization. She focused on researching group cognition models and initiated the Withered Sea Project to conduct simulation experiments, only to fall into a mental crisis as a result. Despite her warnings triggering controversy and even leading to her exile, her research was inherited by her student, Liang Moqing. Amid the celebrations of victory, Yu Baiyun’s reflections on the heart of civilization’s exploration became an important issue hanging over the future of the Xinglong civilization.

This AI short film is adapted from the short story “The Withered Sea Project” from the Computational Universe.

Tools

This time, video generation mainly involves Kling, Hailuo, Vidu, and Higgsfield, while for image generation, I’m relying on old friends Flux and Midjourney, along with Vidu’s multi-reference feature. In video generation, Hailuo is the main tool for card drawing, handling all shots except for close-ups of characters. Hailuo is very stable and occasionally provides surprises. For some particularly exaggerated character emotions, Hailuo is also needed, though in some cases it can be a bit too intense.

For medium shots of characters and beyond, Kling is used, especially since “The Withered Sea Project” involves many dialogue scenes, so KeLing has also drawn quite a bit of inspiration. I mainly use Kling 2.1 Master Edition for generation, which can retain facial textures —a crucial detail for me, as I prefer realistic faces. Kling’s character expressions are restrained, but upon closer inspection, they become more captivating.

Vidu is used for its multi-reference function, which allows specific characters to be placed in designated scenes, which is very useful. The downside is that sometimes the character proportions become skewed, and there’s blurriness. Even in Q1 1080p resolution, Vidu’s blurriness is still visibly noticeable.

Higgsfield is used for some special effect shots. The effects it generates are still quite powerful, but it is expensive, so I only use it for special effects and not for other tasks. Its image generation feature is still quite a bit behind Midjourney. Midjourney remains the main tool for image generation.

This is a rather complex topic because uniqueness doesn’t necessarily guarantee traffic or public recognition. Uniqueness is more like a faith recharge. It means you genuinely believe that your work must have uniqueness, that your personality must be expressed, regardless of the final outcome. If you believe this, then you’ve achieved the first step.

This is part of my 25 short film series; the first of them all. It is adapted from my short story, “The Resolve.”

“The Resolve” depicts Shu Ying’s solitary struggle on the planet Jinqiong. Driven by her yearning for the starry sky, she repeatedly defies the “bowing down” rules of the Jinqiong civilization. Her acts of rebellion provoke a fierce response from the Jinqiong civilization, ultimately leading to her exile to the planet’s surface. On the surface, she faces ruins and solitude, yet continues to pursue her beliefs and freedom.

The greatest appeal of writing genre fiction for me is world creation. I’m part of the generation that grew up playing games. For instance, I’ve been a World of Warcraft player since the open beta, and I still occasionally participate in role-playing because its worldbuilding and culture are so vast. Translating the official World of Warcraft novel Of Blood and Honor was my first formal encounter with writing.

There’s a quote I really love: “I was born too late and missed the Age of Exploration. I was born too early and won’t live to see humanity explore the brilliant stars.” Is this AI wave a great voyage of discovery, or a renaissance? It has given my creativity a chance for complete liberation. For a while, I was troubled by my slower progress compared to others, so I made an effort to form a team; however, in recent months, I’ve come to terms with it.

This is my research direction, and I specialize in realistic long-form narrative. Although I don’t currently have a complete series of work, I’ve spent 1,882 hours on AI filmmaking, with most of that time dedicated to researching the possibilities of AI in long-form narrative. There are already dozens of minutes of AI series content, so long-form narrative is not impossible—it’s just a matter of how well it’s executed.

Without realizing it, a full year has passed since I started creating AI videos. Although most of the time was spent mastering the technology rather than producing a sufficient number of works, and I didn’t successfully form a team, starting from this new year, I will focus wholeheartedly on creation and continuously output works.