With Jay AI Videoof bursting into flames.Teach you to make the same first and last frame of Jay Chou's AI video, AI generated "memories kill" to record his growth course.Like this one with "AI first and last frames"The breakout video of doing a 'life replay' caught fire along the way.
In the past two days, someone used AI to make time-reversal videos with 20,000 likes on one video!
Compared to the normal growth of a person, a reverse playback like this one from aging to youth. It carries more of a magical yet yet real impact. This recollection of the time change of family and friends, the lamentation of the passage of time, and the awe of life itself will be more touching to the audience.
It's really 爷青回..
Think about how you can use this AI video as well. Letting parents see themselves back as adults or even children. I would also feel very emotional.
It just so happens that there are people begging for this kind of tutorial, and that the techniques used for this kind of video can actually be accomplished with many AI video tools, whether it's Korin, i.e. Dream, Vidu, etc.
But here's a little more information on what first and last frames are.
First and last frames, as the name suggests, are used to control the generation of a video with a first frame and a last frame. The middle clip is the result of the AI generation, which can slowly shift from the first to the last frame.
It may seem like it's quite complicated, but it's actually ultimately simple to have hands.
Before we get started, let's first open Deepseek to generate the cue words.
The website is here: https://chat.deepseek.com/
Enter this prompt in Deepseek and click Send.
Generate me a painting prompt for each of the 5 ages of a Chinese man, from old age to youth. Each image is a full body shot and is doing a representative action. Just output me the cues for the paintings.
Once it's sent, Deepseek outputs this cue for each age group to us, and we just copy the cues that
Then come to the i.e. Dream AI: https://jimeng.jianying.com/
Click on AI Generation, copy in the cue words for the drawing, select the model for Instant Dream 3.0 Pro, and click Send.
A 70-year-old Chinese grandfather, with a full head of gray hair and a kindly wrinkled face, dressed in a traditional Tang suit, is playing tai chi in the park, his movements slow and powerful, with the soft sunlight of a bamboo forest in the early morning in the background.
This way we generate images of older people. Of course, we can also generate images of middle-aged people and young people. After generating them, click on Generate Video here and select the first and last frames here.
The first frame can be uploaded for older people and the last frame for middle-aged people.
Then enter the cue word for men slowly getting younger, this way we have this video that generates this video clip from an older man to a middle-aged man.
Similarly, we can also upload the middle-aged man in the first frame, and the young man in the last frame, and then input the prompt words of the man who is slowly getting younger, then we can generate this video clip of the middle-aged man to the young man.
Then after these video clips are generated, finally come to a cutout, we can add a voiceover to the cutout, splice these video clips together, and a finished product is generated.
Of course in addition to the father's version, the mother's version can be made as well.
Elderly grandma:
Middle-aged women:
Young girls:
The normal world is from young to old, but in the world of AI, time can really be rewound to experience the feeling of growing up in reverse.
It doesn't need a very complex plot, simply by inverting the timeline, it is able to speak to the common human emotion of time and life that
It's only logical that this AI video would become a hit.