If you were to create this kind of effect, you needed to film at least twice and lots of editing.

@aisavvy

131.1K followers · usually 4.1K views a post

Views

57.7K

Likes

171

Comments

20

Shares

11

SoundTikTok sound · NICE VIDEO

Promoted product

OpenArtopenart.ai

Create AI art, images, videos and music online for free. 100+ models from Google, OpenAI, Seedance & more. — all in one place. Trusted by 8M+ creators. Start in seconds.

21 videos7 creators

Why it worked

The video uses a high-contrast visual hook (the transformation effect) to immediately capture attention, followed by a clear, actionable tutorial that solves a specific creative problem.

Summary

The creator demonstrates how to use the OpenArt AI VFX tool to transform a simple video of himself lying on the ground into a more complex visual effect. He provides a step-by-step tutorial on masking and using reference images to achieve the desired output.

Structure

1Identify the problem with traditional editing
2Introduce the AI solution
3Step-by-step screen recording of the tool

Original caption

AI VFX made easy #aivfx #vfx #openartvfx #openart #aitools

More from @aisavvy

Product placement

OpenArt AI is the main subject of the video, shown through a screen recording of its interface.

Call to action

If you comment hoodie, I'll send you my prompt.

On-screen text

AI SAVVY to effect, Now one AI tool First record a video of yourself laying on the ground Dogs are optional. openart.ai VFX maximum duration is 8 seconds. mask the area that you want to stay unchanged. confirm reference image create it with AI model is GPT Image 2 write your image prompt hoodie resolution generate

Transcript

If you were to create this kind of effect, you needed to film at least twice and lots of editing. But now, you just need one AI tool and a bit of imagination. First, record a video of yourself laying on the ground, filmed from above. Dogs are totally optional. Next, go to OpenArt AI, select OpenArt VFX, and upload your clip. The maximum duration is 8 seconds. Then, mask the area that you want to stay unchanged. In my case, it's just the head. Once you're happy with the mask, click confirm and add a reference image to guide the final look of the video. You can either upload one or create it with AI. Select the image model, in my case it's GPT Image 2, and write your image prompt. And if you comment hoodie, I'll send you my prompt. Then, optionally describe your video idea, select the resolution, and click generate. So, give this idea a try and let me know what you think.