Guide to Deforum in Automatic1111 Yeni, March 7, 2024March 7, 2024 Content What is Deforum? Deforum Without Complicated Installation How to Use Deforum to Generate Animation Deforum Settings Demo What is Deforum? Deforum is one of the tools in Stable Diffusion to create animation. It use Stable Diffusion’s image-to-image function to generate a series of images, which are then stitched together to form the video. Furthermore, Deforum makes minor changes to an image frame and then uses the image-to-image function to generate the following frame. Since the difference between frames is tiny, it produces the illusion of a continuous video. Deforum Without Complicated Installation DiffusionHub makes deforum easily accessible for you since you don’t need to do the installation. All you need to do is put in the prompt and configure the settings to adjust the “camera” movement. To help you with the camera movement, there is a tool that helps you to test it called Deforum Helper. The tool is free and has many uses as below: Generate visual timelines for your Deforum animations. Set and calculate the duration and FPS of your Deforum project. Effortlessly edit animated values using a dynamic editor. Experience instant, real-time 3D visualization of your animations. How to use Deforum to Generate Animation? Before we delve deeper into deforum, it is important to understand some settings in the Deforum tab. Essentially, there are parameters you need to set below: Go to Deforum Tab in DiffusionHub. 2. On the Keyframes Tab, check the following parameter: a. Animation Mode: Choose which mode you want your animation to. There are 2 options, 2D & 3 D. Generally, 2D mode is series of images as 2D and perform modifications such as the magnification and rotation to create movement illusion. In the other hand, 3D mode is series of images as a perspective on a 3D scene. You can change the camera’s viewpoint in any 3D action. b. Max Frames: This is to determine how long do you want your video duration. The bigger the max frames, the longer the video will be. c. Motion Settings: 2D Motion Settings Angle: In 2D, it will rotate the image. A positive value rotates the image counterclockwise, and a negative value rotates the image clockwise. A larger value rotates the image faster.By default, the rotation is around the center of the image. You can control the center of rotation by setting Transform Center X and Transform Center Y. 2D Translation X: Use Translation X to move the image sideways. Use a positive value to move the image to the right and a negative value to move the image to the left. 2D Translation Y: Use Translation Y to move the camera up and down. Use a positive value to move the image down and a negative value to move the image up. 2D Transform Center: Transform Center is for changing the focal point of zoom and/or rotation. 3D Motion Settings 3D motion serves as an alternative to 2D motion. Imagine holding a camera and freely moving and rotating it in any direction. 3D Translation X: Move the camera sideways using the Translation X parameter. Positive values move the camera to the right, while negative values move it to the left. 3D Translation Y: Adjust the Translation Y parameter to move the camera up and down. Positive values move the camera up, and negative values move it down. Translation Z: Similar to zoom in 2D motions, Translation Z controls the zoom effect in 3D. Rotation X, Rotation Y, Rotation Z: These parameters allow you to rotate the camera around the X, Y, and Z axes, respectively. Perspective flip: Perspective flip performs 3D-like transformations to the image to create some cool effects. You will need to select Enable perspective flip to enable these options. Motion Schedule Motion settings in Deforum are defined using the following format: frame1:(value1), frame2:(value2), frame3:(value3) etc. Each item consists of two integers separated by a colon: the frame number at which the motion takes effect, and the matching value contained in brackets. Dynamic visual effects can be achieved by combining and modifying multiple motion parameters over time. As mentioned above. Deforum helper app is helpful tool to visualise the camera movement for your animation. 3. Prompt Writing Prompts here require JSON language. It is quite helpful to use JSON validator to make sure your JSON prompt has the correct settings. Moreover, please note the max frames you input previously in the keyframes tab. As you see above, there are “0”, “30”, “60”. This means from frame 0 to frame 29, it will use the “0” prompt, and at the 30th frame, it will switch to the “30” prompt. Deforum Settings Demo Now let’s play around with the prompt and settings in Deforum for beginners: Don’t forget to set the checkpoint for your base model. Each base model produce different images so if you would like to have a certain styles, choose which models you like. In this case, we choose RevAnimated to generate anime looking video In this tutorial, we set it into 3D mode and 120 Frames max You can also follow the template motion settings & motion schedule settings below: Translation X: 0:(0), 30:(3), 60:(3), 90:(0) Translation Z: 0:(0.2),60:(5),90:(7) Rotation 3D X: 0:(0),60: (0),90: (0.5) Rotation 3D Y: 0:(0).30:(-3.5).60:(0.5).90:(-2.8) Rotation 3D Z: 0:(0),30:(0.2),60:0 ,90:(-0.5) 4. Prompts: You can play around with prompts, however, make sure that it is on json format. It is quite helpful to use json validator website. Once you type your prompts, click on generate. After the generation done, don’t forget to click “click here after generation to show the video” Voila! Your video now is available and can be downloaded at /dh-user/output/img2img-images/Deforum See how easy it is to create animation in DiffusionHub? Head over DiffusionHub and let’s create some more! Share on FacebookPost on XFollow usSave Uncategorized
Uncategorized Understanding LyCoris and LoRA April 1, 2024March 24, 2024 Hello again fellow Stable Diffusion Enthusiast! Today, we are going to explore LyCORIS models – and no, it’s not an alien language, it stands for Lora beYond Conventional methods, Other Rank Adaptation Implementations for Stable diffusion (I know, sounds like a mouthful, but bear with me, it’s all fun!). So,… Read More
Generative AI Introducing HelloWorld: a realistic SDXL base model February 7, 2024February 14, 2024 Table of Contents: HelloWorld Evolution It’s 2024 and AI photography has evolved so much compared to the last year. The new SDXL model HelloWorld marks a groundbreaking transition from the established SD1.5 platform to the futuristic SDXL series. This new SDXL base model series not only symbolizes a departure… Read More
Uncategorized Stable Diffusion 3 April 29, 2024April 29, 2024 This week, Stable Diffusion community has been buzzing with the release of Stable Diffusion 3 API. Just what is exactly difference between the previous model and SD 3? We’ll deep dive into this blog post! What is Stable Diffusion 3? Stable Diffusion 3 (SD3) is an advanced text-to-image generation model… Read More