I want to generate new videos based on the attached video example. Just like DALL-E makes with images.
For example, I have a 20 seconds video with me and my friends running on botanical garden.
The new generated video, must use the same elements from the video, also must to change any kind of copyright, like faces of people, t-shirts (some T-shirts worn by runners have the name of the athletics club they belong to written on them), etc.
The Most important thing: AI MUST RECOGNIZE PERFECTLY ALL ELEMENTS FROM THE VIDEO !!
This is a video example:
https://www.youtube.com/shorts/bYVIQdU9xJY
---- Example prompt for regenerate video----
I want a sporty and motivational background music to complement the video, without voice or subtitles. I will include this preference in your video project setup. Before generating the project, let’s summarize the details to make sure everything is in line with your vision.
Duration: Video of maximum 20 seconds
Scenario: Running with friends in the botanical garden, capturing the energy and connection between participants in a natural setting. We walk up some stairs made of marble, there are many flower arrangements and trees around. AFTER walk up the stairs, we were running in a straight line on a cobblestone corridor, with lots of sun. There are many trees lined up in a straight line, to our left and to our right
Tone: Energetic and motivational, suitable for an audience interested in sports and outdoor activities.
No Voiceover or Subtitles: There will be no narration or subtitles in the video.
Background music: A sporty and motivational track to inspire and energize, avoiding house or trance genres.
If these details match what you want, tell me “Continue” to generate the video project. If you have any changes or additional details to add, please let me know!
I am interested in developing an AI feature that can revolutionize the process of toasting various food items to achieve perfect browning. Leveraging cutting-edge AI vision, this feature should meticulously detect the level of browning on products like toasts, bagels, English muffins, and pastries in a toaster or toaster oven, as well as employ AI-controlled tools to interact with the toaster.
The success of this innovation would rest upon the AI’s ability to ACCURATELY AND CONSISTENTLY IDENTIFY THE BROWNING LEVEL on various products within the toaster!
---- Example prompt for AI toast feature----
I’d like this advanced AI capability to enable complete customization in terms of the desired level of browning for different items, catering to individual preferences while making toasting a truly efficient and personalized process. The main considerations for your AI model are:
Visual Accuracy: The AI vision must be sharp enough to discern variations in the browning level and differentiate between diverse products and their specifications.
AI Controlled Tools: The AI must coordinate with suitable equipment to interact with the toaster or toaster oven. It should maintain user safety and guarantee an unobtrusive, practical use experience.
Interface & Interaction: The AI functionality should integrate seamlessly with contemporary kitchen appliances. It could provide easy-to-use and interactive options (including smart devices) for users to set their preferences.
Adaptability & Learning: To achieve optimized toasting, the AI must be capable of adapting and learning from user preferences over time and effectively adjusting the process for desired results.
If this concept aligns with your vision for AI-driven culinary advancements, say “Proceed” to initiate the development of this state-of-the-art toasting technology. Should you have any modifications or suggestions, please feel free to share them!
this is a video example, from which I made the scenario. Important is AI to recognize the element from the video, and make something similar, with the same elements, but also different.
You can do this very easily. Let’s say my video is 1 minute long. The AI will have to stop the film from place to place, let’s say every 10 seconds, and make a print screen, i.e. a frame. So in total there will be about 6 frames.
Then the AI will identify all the elements in each frame separately, it will realize the differences in the images, and it will know how to create a video from the frames placed one after the other.
At the same time, the AI can additionally ask for some indications about the film proposed by me (attach) to compare with the elements identified in the frames.
I say that it is easy to make such a film, from frames. At least generate a new film with a duration of 30 seconds, a remake after the original video.
deo