General Introduction
Dreamina is an AI image creation tool launched by Jitterbug, designed to help users generate images using text descriptions. It is based on a large multimodal model developed by ByteDance, which can transform simple text into wonderful images.Dreamina features include text-to-image generation with advanced editing options such as HD map/expanded map/local redrawing, which is suitable for all kinds of users, from professional artists to amateurs, who can easily get started with the process and turn their imaginations into realistic visual works.
Dreamina Reviews
vantage
The advantage of Dreamina is that it can generate multiple dimensions and styles of images based on the text entered by the user using AI technology to meet the different needs and preferences of the user. Users can choose different image sizes and template types, as well as save or re-generate images or edit them more on CapCut.Dreamina helps users create great images with simple prompts, whether they are editing photos, videos or GIFs, all of which can be done on CapCut.Dreamina can be used for Jitterbug's graphic or short video content creation, enriching Jitterbit's content library in AI creation.
drawbacks
The disadvantage of Dreamina is that it is still in the internal testing stage and requires an invitation code to experience it, which limits the scope of users. In addition, the images generated by Dreamina may not necessarily meet the user's expectations, or there are some quality or copyright issues, which require the user to make their own judgment and adjustments. Last but not least, Dreamina's function is still relatively single, it can only generate images based on text, there are no other AI creative tools, such as video, music, text and so on.
Discussion on Dreamina
Dreamina's website reviews are mainly from a number of media and user reports and comments, here are some excerpts:
36 Krypton believes that Dreamina is an attempt by Jitterbug in the field of AI creation, which can provide more materials and inspirations for Jitterbug's graphic or short video content creation, and also attract more users and creators.
- AIBetas said Dreamina is the first AIGC product from the Shear Image team, demonstrating Shear Image's strength and innovation in AI technology, and providing users with a fun and convenient tool for generating creative images.
One user commented on Weibo that Dreamina is a fun tool to generate various styles of pictures according to your own ideas, and sometimes it's fun to see some unexpected effects.
Another user replied on Zhihu that Dreamina is an amazing tool to create stunning visual effects with words, but there are some drawbacks, such as the quality of the generated images is not high enough or there are some repetitions and similarities, which still need to be improved.
web competition
Competition for Dreamina's site comes mainly from a number of other AI creative image generation tools, some examples of which are listed below:
- [Artbreeder]: A tool that can generate and mix various types of images with AI, including people, animals, landscapes, styles, etc. Users can adjust various parameters and options, or upload their own images for editing.
- [Deep Dream Generator]: a tool that can convert images into fantastic artwork using AI, users can choose from different styles and filters, or create their own, or participate in the community to share and communicate.
- [This Person Does Not Exist]: a tool that can use AI to generate images of non-existent faces, with a new face appearing every time the web page is refreshed, which users can use to test their ability to recognize faces, or for some legitimate use.
Dreamina Help
Enter the copy you want to convey or choose a prompt word template
Refinement of work using editing functions such as HD drawing/expansion/partial redrawing
View the generated image and make necessary changes
Save or share to various platforms
Dreamina Feature List
Generate images from text descriptions
Supports HD map, expanding map and local redrawing, etc.
Provide prompt templates to simplify the creative process
Easy and fast AI drawing creation
Suitable for a wide range of content creation, such as social media sharing, project presentations, etc.
Dreamina is easy to use
1. Enter the prompt, referring to the writing conventions of Stable Diffusion prompts.
Simple geometric shapes made of colored ribbons, lines filled with halos of light, azure blue, soft cool colors. Cinematic, epic, realistic, 8k HD, wallpaper. In the Model and Scale options you can simply control the image style and quality.
2. Matting to generate images containing matting elements
Import the reference image, you can recognize the subject in the reference image or the face wrapped in the image, here we generate the wallpaper, I would like to generate the image to have a kind of wave scales in the sea style (actually not feasible).
Note: If I upload a pristine sea or wave it is not recognized as a subject or face, so the current imported reference image is not suitable for generating a scene with no subject item or a face that cannot be recognized. To demonstrate, let's force in an image that contains a subject.
As we can see, the image generated by adding the recognized subject image to the cue is well defined, which is an example of a bug, from which you can tell that Dreamina's "matting" feature, currently described as "inserting" faces or objects into the image, is more appropriate.
Example of the correct use of reference diagrams in Dreamina
3. Detailed repainting
Getting a more detailed image, this feature compares to Stable Diffusion, which is not actually hand and face optimization and does not enhance the current image quality. It's impossible to accurately judge the usefulness of this feature at this point, it looks like a random draw with basically no change to the subject.
4. Generate a single image after the image editing function
Local repainting: adding a big moon
Using the Eliminator Pen: Remove the large tree on the right side
Expanded and Ultra-HD images:
Final product: Comparison between the original and the final product
Currently, the Ultra HD feature does not complement and repair image details, especially the redrawing ability is limited, as of February 2024 version of this feature is more ribbed. In use you need to understand a basic concept: image enlargement, full image redraw, full image detail optimization, repair, color filling, photo restoration, will be implemented using different technologies, to clarify the purpose of your use, and then evaluate whether the tool meets expectations.
5. Video Generation
Support Image to Video and Text to Video, each time to generate video will consume 12 points, the duration of the generated video is 3s, if you are a VIP, the maximum can be extended to 9s.