TapNow Seedance 2.0 Real-world Test: AI-powered creation of "Brands" and "Paper Phones" - Full Process Open Source

These days AI Numerous masterpieces of video content have emerged.

For example, the visual masterpiece "The Sign," which blurs the lines between reality and dreams, has garnered over 50 million views across the internet and has become a hit on Bilibili. AI The winner of the creative competition received a million-dollar prize. Even Guo Fan, the director of "The Wandering Earth," shared and liked his post.

For example, the short film "Paper Cellphone," which moved the entire internet to tears, had no real actors involved and was created by just two people in three days, yet it touched the hearts of tens of millions of netizens.

You might think these are only for tech gurus, but actually, the complete workflow for these high-quality, viral videos has already been implemented.open sourcePublished on TapNow, the creation process and core elements of these viral videosPrompt wordsWe can learn directly from it.

Recently, TapNow launched Seedance 2.0 and launched an unprecedented global creative competition called "Ten Thousand Parallel Universes".

Are you ready to replicate the next blockbuster product with tens of millions in sales? Let's get straight to the point.

We open the TapNow website.

Official website:https://www.tapnow.ai

TapTV has countless top creators.open sourceof AI Video workflows, including TV commercials, animations, narrative short films, and music videos, are all available.

By hovering the mouse over any short film, you can preview the video and view the creation process.

for example AI Live-action short film "Paper Cellphone":

Just like shooting traditional films and television dramas, doing AI The first step in making a short film is casting.

AI actor

TapNow comes pre-loaded with tens of thousands of copyright-compliant live-action videos, ready to use right out of the box. The content is incredibly rich, ranging from the ancient charm of the pre-Qin period to the elegance of the Ming and Qing dynasties, from the Renaissance to the cyberpunk future; even alien species and deities, spanning ancient and modern times, both Chinese and foreign. If you can imagine it, it has it.

These characters all supportOne-clickApply it to the canvas. Even if you have absolutely no experience, you can immediately have this huge cast of actors.

If you want a unique protagonist, you can also create your own character by customizing their appearance.

In the TapNow canvas, we add a "Text-to-Image" node. Using TapNow's built-in model to generate live-action images, no enterprise verification is required to directly generate videos with Seedance 2.0. (For content uploaded by yourself or generated from "Text-to-Image," enterprise verification is required to enable the live-action verification channel.)

We create AI The actor can try double-clicking on a blank area of the canvas to add an image node.

enterPrompt wordsGenerate a portrait image using image nodes.

Click on the role compliance verification above the image. After a few seconds, a blue icon will appear on the Image above the image, indicating that the verification was successful.

After verification, our AI Actors can then appear directly on camera.

For example, we used Seedance 2.0 to arrange an audition for her:

Create a video node and connect it to an image node, adding the image as a reference image.Prompt wordsmiddle.

Generate an audition self-introduction video of the girl in the image (@), against a pure white background. The girl stands facing forward, smiling naturally and confidently, and introduces herself: "Hello everyone, I'm K-sister, 18 years old, 163cm tall, and weigh 82kg. My works include 'K-sister Research Society,' and my hobbies are singing and playing the piano." She then shows her left side, back, and right side before turning back to face forward.

Let's take a look at the generated video together. The character's movements and postures are completely consistent.Prompt wordsRequirements.

Try adding a little more emotion, let AI The actor conveyed five different emotions in a continuous performance:

Referencing @Image's image of a girl, against a clean white background with soft front lighting, the girl faces the camera and expresses five different emotions while saying "I don't want to." 0-3 seconds: Coy mode – The girl pouts, tilts her head, sways gently from side to side, and shakes her fists in front of her chest, saying in a coquettish tone, "I don't want to~" with a rising and drawn-out ending, her expression playful and cute; Hard cut → 4-6 seconds: Angry mode – The girl frowns, her eyes are sharp, she crosses her arms over her chest, leans slightly forward, and shouts forcefully and sharply, "I don't want to!" with a strong and forceful tone; Hard cut → 7-9 seconds: Sad mode – The girl's eyes are red-rimmed, her gaze is downcast, her lips tremble slightly, she pushes her hands forward weakly, and her voice is choked and trembling. Trembling, she says, "I don't want to..." slowly, with a sob-like tone at the end; hard cut → 10-12 seconds: resolute mode, the girl raises her chin, looks directly at the camera, her eyes calm and firm, raises her right hand with her palm facing outward in a "stop" gesture, and clearly and forcefully says, "I don't want to," without the slightest hesitation; hard cut → 13-15 seconds: fear mode, the girl's eyes widen, her body instinctively shrinks back, her hands cross over her chest, and she says urgently with a trembling, sob-like voice, "I don't want to!" while slightly turning her head to avoid eye contact. The five segments are sharply cut, each lasting 3 seconds, with distinct facial expressions and body language.

Look at the flow of these emotions, isn't it incredibly natural? It feels so real, like someone who's truly alive.

AI All-around director

With the actors secured, the next step is filming.

I've done AI As anyone who's been through the video game knows, generating videos often involves drawing cards, which has a high rate of discarded cards, and going through them one by one is incredibly frustrating.

To address this industry pain point, TapNow recently quietly launched... Agent Function, is a AI All-around executive director.

We just need to send the inspiration to AgentIn conjunction with Seedance 2.0, Agent They will handle the rest of the content generation work.

We give directly Agent Issue the instruction:

Using Seedance 2.0, generate a 30-second short film titled "Some Handsome Guys, You Can See Them, But You Can't Have Them" based on my storyboard. The main character is @Image1; Handsome Guy 1 is @Image2; Handsome Guy 2 is @Image3; [Script Content]

After receiving the instruction,Agent They will then begin confirming information and creating content. They will proactively contact me to align the production plan and materials, and then we will begin production at lightning speed.

final Agent Two 15-second clips were generated, which I manually spliced together, and a short film was completed:

The characters look very realistic, and their emotional changes are also very well done.

We can even let Agent Generate dozens of alternative shots with different camera movements and micro-expressions at once.What we need to do is, like a true master director, pick out the most stunning scene from a pile of high-quality finished films.

Previously done AI Video editing is like playing a game with tools. We have to constantly guess which approach the model will work with.Prompt wordsI'm guessing if I'll succeed in this gacha pull. While not every pull in TapNow is perfect, Seedance 2.0 certainly reduces the cost of repeated trial and error.

More importantly, TapNow gives more and more creators the opportunity, for the first time, to bring the world in their minds to life.

Having come this far,AI It's more like a paintbrush that gets easier and easier to use, helping us shorten the distance from inspiration to finished product, and giving more ordinary people the opportunity to enter the world of video creation for the first time.

But what truly gives a work its value is always the person behind the paintbrush. How you see the world, why you want to tell this story, and what emotions and meanings you want to convey to the audience—all these things will ultimately find their way into the artwork.

Perhaps the next image that will touch the world is hidden in your mind.

Original link:After the viral success of "Brand" and "Paper Cellphone," tens of millions of copies were sold. AI Video Workflowopen sourceAlready