Some weeks ago, we released a text-to-image capability (powered by Ideogram): users are able to generate images directly from the prompt space, or even modify attached images.
We want to provide even more capabilities: now that image generation is better integrated into the prompt space, we can use those images to create short videos. This new image-to-video capability is powered by Runway Gen-3, and it’s fully integrated with Straico’s platform.
Before Starting
Click on the bulb-icon on the right-hand side of the prompt space: this will open the capability selector:
Make sure that the capability “Image to Video (Runway)” is active in the capability selector.
Remember that this capability is only compatible with some LLMs (like the ones from OpenAI).
Generate
There are two ways to generate a video from an image in Straico:
From an attached image
If you have the following image in your local files:
You can upload an image from the attachment section:
Once the image has been put into the prompt space, you can send a prompt to generate a video inspired in the uploaded image:
“use the attached image to generate a video, 5 seconds long, of the fox running along the shore”
The resulting video also appears in the prompt space:
The resulting video looks like this:
From a previously Ideogram-generated image
You can generate videos from images previously generated with Straico’s image generation capability.
Make sure that the image generation capability is turned on. Then, send a prompt to generate a new image:
Once the image is there, you can send a prompt to create a video inspired on the recently created image:
The resulting video looks like this:
Parameters
You can specify the following parameters in the prompt for generating a video from an image:
- Ratio: the aspect ratio of the video, 1280×768 (landscape), or 768×1280 (portrait).
- Duration: You can set a value between 5 seconds and 10 seconds. Beware: longer videos will cost more coins
Recommendations
Image-to-video capability is not cheap: use it wisely. Make sure of having a good quality image to start, and try to make your prompts as specific as possible. This way, the resulting videos might have better quality.
Conclusion
Straico’s new image-to-video capability is a new step for integrating several AI-related tools into one single platform. We’d love to know how it works for you, and we hope this new feature is an ally of your creativity.
