I FIND THAT A LOT OF PEOPLE DON'T KNOW HOW TO USE AI, BUT THEY DON'T KNOW HOW TO USE IT
Especially after the dream of Seedance 2.0 comes out, making video is really too simple. Now you're going to the big platforms and you're going to find out that the quality of AI's video is no longer the same: it's clean, it's very cool, it's very cool, it's full, it's big, it's like a big, it's like an IA micro-film, it's got big, it's got big, it's got big, it's got big, it's got big, it's got big, it's got big, it's got big, it's got big, it's got real, it's got really real 3D country。
NOW THE AI VIDEO RING, THE TECHNOLOGY HAS GOT TO THE CEILING. BUT DID YOU FIND A PROBLEM? ALTHOUGH WE'RE ALL DOING ALL KINDS OF AL VIDEOS OF FAIRY FIGHTING, THERE'S HARDLY A LOT OF IDEAS THAT REALLY COME OUT OF PURE “IMAGINATIVE” AND “TALENT” TO GET PEOPLE TO READ “HOW CAN THEY PLAY LIKE THIS”。
When all people can easily make high-quality videos, simple visual stimulation is difficult to keep。
Today, I'm going to give you three ideas to teach you how to break the mindset and make your AI video imaginative. We don't need any complex nodal operation, but we need to use the core of the "tribe" to complement the most burning tools of the moment (Nano Banana Pro, dream, coven, conch, Midjourney, etc.)。
First line of thinking: Breaking the laws of physics -- rewrite the rules and create visual wonders
Everybody's doing it AI VideoAnd when it comes to pictures, there's often a subconscious habit: to make it more "real."。
The light must be natural, gravity must fall, and the object must be stabilized. But this subconscious pursuit of “reasonable” behaviour is the main reason why your imagination is trapped。
YOU HAVE TO UNDERSTAND A LOGIC AT THE BOTTOM OF COMMUNICATION: CREATIVITY OFTEN COMES FROM IRRATIONALITY. THE VIEWERS IN THE SHORT VIDEO STREAM ARE NOT ATTRACTED TO “RESOLUTION”, BUT TO “CONFORMITY”. NOW THAT WE'RE IN THE VIRTUAL WORLD CREATED BY AI, YOU CAN TOTALLY "REVERSE IT." AS SOON AS A FEW PHYSICAL RULES WERE BROKEN, THE PICTURE WAS DIFFERENT。
The core of the exercise and tipping down:
We break down the laws of physics from two different dimensions:
Dimension One: Reversing Gravity - To make water flow up, everything lose weight
This is the most direct and visual impact. It's the opposite of what's originally affected by gravity. Let's look at this classic case。
Case one: Antigravity polar bear and Fallback Falls
Look at this. If it's just a polar bear sitting in front of a forest waterfall brushing a cell phone, it's just a normal “work-out” map. But what if we change the laws of physics
Take a closer look at the details of the picture: the giant rock with moss beneath the polar bear is not attached to the ground, but is completely floating in the air. What is even more exaggerating is the waterfall in the background, where the water flows, far from being dumped, go against the gravity of the Earth's core and rewind into the sky with a huge column of water. The surrounding gravel and beads were also deweighted and floating in the air。
This sense of super-realism makes the brain of an audience impervious, generating a strong visual presence。
1. Static graph formulae:
Formula:
Subject + General Environment + Paranormal Physical State (core) + Superrealistic Style Word + Film-class Shadow
Prompt Demonstration:
Chinese:
class="prompt" film-level real photography. a polar bear with hairy hair and black coarse collars is looking with two claws on his smartphone. extremely convulsive superrealism: polar bears sit on a huge rock with moss, but it is floating in half the air. the background is a green forest, a huge waterfall that runs against the gravity of the earth and the current is roaring backwards into the sky. there are heavy water beads and rubble floating in the air. 8k resolution, very detailed。
English.
Highly detached Poland, intuitively looking at a smartphone in its wakes.
Video generation techniques (i.e. dream / coven / conch):
USING AI TO GENERATE THIS COMPLEX ANTI-PHYSICAL DYNAMIC, THE BEST WAY TO DO SO WOULD BE TO USE “CHARACTER VIDEO”。
Imports a static map as a reference (prime frame) into a dream or a ghost。
Video alerts need to be restrained and focus only on environmental dynamics:
Keep the picture weightless. The water flow from the waterfalls in the background continues to flow up and down into the clouds, and the debris and beads in the air slowly float up and down. The polar bears on the rocks remained stable and suspended and their thumbs slipped gently on the mobile screen。
THIS WILL MAXIMIZE THE RISK THAT AI VIDEO MODELS FORCE "CORRECT" IMAGES BACK TO NORMAL IN MOTION (E.G. IF THEY SUDDENLY FALL OFF A ROCK OR LET THE WATER FLOW DOWN)。
2D: The paradox of mass and size -- slightly floating giant buildings
Something that is extremely heavy and that should be firmly connected to the ground can produce another huge visual shock and sci-fi if it displays a light, floating posture。
Case II: A small floating giant
Imagine a skyscraper that is high in the clouds and that does not connect to the ground at the bottom, but floats slightly and slowly, like a weightless balloon in the clouds. The bottom can even see broken steel bars and falling mud。
It does not need to be as violent as the polar bear's fallback waterfall, but a sense of “light” discord can create a very powerful narrative。
Prompt demonstration (for all tools):
Chinese:
film-grade science-fiction, cyberpunk city vision, at dusk. the huge steel skyscraper complex on the horizon did not connect to the ground, but drifted slightly and slowly up and down in the clouds, like a weightless balloon. at the bottom, there are complex pipes and broken steel bars. film-class photo, 8k resolution。
copy
English.
Cinematic sci-fi photography, cyberpunk city vista, Sunset.
Remember, in the world of AI, stop chasing "like a real shot." Add the Anti-gravity (retrogravity), Levitating/Floating subtly (floating/light suspension), Flowing upwards (upwards), the moment when you rewrite the rule, the moment of creativity。
Second line of thinking: making time out of control
MOST PEOPLE HAVE A VERY SINGLE UNDERSTANDING OF THE "TIME" TREATMENT IN THE AI VIDEO. MOST OF THE OPERATIONS ARE SLOW, FAST, CARD POINTS。
But it's just a "change in the rhythm" of the video, which is not creative. The real impact of time is to make time “problems”. When time is abnormal, the mind of the audience is instinctively alert and attracted to death。
The core of the exercise and tipping down:
1. Generating a sense of spatial and temporal fragmentation (with different speeds of time running between the protagonist and the background)
Lets part of the picture pass and the other part stops。
Graphical induction techniques: The emphasis on background “motion blurry” and the subject “absolute static”。
Nano Banana Pro / Midjourney:
Chinese:
Hee-hee crossroad. The main character in the center of the picture is looking down at his watch, and his body and clothes are wrinkled and completely static. And the pedestrians and vehicles around him were in a state of extreme ambiguity, as if the surrounding time was moving at 100 times faster. Strong time deviation。
2. Unnatural disruption of movement (half of motion, sudden stop)
Video Dynamic Indices:
Strong sense of time and space. The men in the center of the picture remain absolutely static, with no movement. The background pedestrians and vehicles travel like fast-forwards, producing severe photolytic lace effects and debris。
Third thought: turn the camera into a role
THE CAMERA IDEA IS A PLACE THAT 901 TP3T PEOPLE CAN IGNORE WHEN THEY DO AN AI VIDEO。
A LOT OF PEOPLE DO AI VIDEOS, JUST USING THE LENS AS A "RECORDING TOOL." THE CAMERA IS ALWAYS OUT THERE, WATCHING THINGS HAPPEN LIKE A BYSTANDER。
But do you ever wonder why the camera can't be a character? Ideas are often not new, but new. A change in perspective makes a dry story creative。
The core of the exercise and tipping down:
1. Large objects coming perspective (humans from ground view of giants / animals)
Change the conventional horizon angle, place the lens in a very low position and give it the identity of a tiny creature。
Phrasing demonstration (for all graphics/video tools):
Chinese:
EXTREME VIEWING, FIRST PERSON VIEW OF AN ANT (POV). FROM THE CRACKS OF WEED AND MUD UP, A HUGE HUMAN SHOE FLOOR IS COVERING THE SKY AND IS ABOUT TO STEP DOWN WITH A SENSE OF OPPRESSION. EXAGGERATED PERCEIVING RELATIONSHIPS, MICROSPECIES, A STRONG SENSE OF GIANT FEAR。
2. Limited physical perspective (the world in the cup)
Put the camera in a place where it would not have been possible to put the camera and give it the perspective of the object。
Phrasing demonstration:
Chinese:
THE FIRST PERSON, PERCEPTION (POV), IS LOCATED INSIDE A CUP FULL OF COFFEE. LOOKING ABOVE THE WATER THROUGH DARK BROWN LIQUIDS, A MAN'S FACE IS APPROACHING A CUP OF COFFEE. THE WATER HAS A SLIGHT RADON, AND THE LIGHT RUNS THROUGH THE LIQUID AND PRODUCES A REFLECTION. VERY NEW PERSPECTIVE。
By adding First-person POV from... or Extreme low/high angle looking at... (extremely perceiving/observing...) at the beginning of the hint, you can quickly subvert the daily perception of the audience。
Summarize:
DON'T GET STUCK WITH "REALISM." IN THE AI WORLD, YOU CAN:
Breaking the laws of physics and creating an anomaly of vision。
Make time out of control and create a sense of vigilance about time and space。
Turn the camera into a role, recreate the story with a new perspective。
SO, STOP LOOKING FOR MORE REAL. IN THE WORLD OF AI, RULES CAN BE REWRITED