Anyscene AIAI Handshake Video
Upload one image with two people and get five seconds of them stepping into one scene and shaking hands.
Drop one photo here, or click to upload
PNG · JPG · WEBP · HEIC · up to 10 MB
Your photo is used only to generate your clip and is never used for training. A failed render refunds its credits.
Definition
What Is an AI Handshake Video?
An AI handshake video is a five-second clip generated from one image of two people, in which an image-to-video model places them in a shared scene, walks them toward each other and ends on a firm, friendly handshake.
Two portraits are enough
The sample starts from two separate studio portraits placed side by side in one image. The model turns them into one bright office lobby where the two meet.
Professional by default
Eye contact, a friendly smile and the same outfits as in the photos. The clip suits partnership announcements, new-hire posts and event promos.
Five seconds, landscape friendly
A side-by-side input is naturally wide, so the result works well in presentations, LinkedIn posts and website headers.
If the two people were photographed separately, crop each to a similar size and framing before placing them next to each other. Matching light and scale make the shared scene more convincing.
Three steps
How to Make an AI Handshake Video
Prepare one image, click once, download.
+Step 1: Upload one image with both people
A photo of the two together, or two portraits side by side in one image. JPG, PNG, WEBP or HEIC up to 10 MB.
Step 2: Click generate
Kling 2.6 renders the fixed handshake prompt in one to three minutes for 40 credits, refunded automatically if it fails.
0:05Step 3: Download the clip
Save the five-second 1080p mp4 for your announcement, deck or post.
AI Handshake Video FAQ
What to know before you generate.
40 credits per clip on a paid plan. Failed renders refund their credits automatically.
No. Use photos of yourself and people who have agreed to appear. A handshake clip can suggest a real-world relationship, so never make one that could mislead.
The model invents a shared setting that fits the photos, usually a bright, neutral space. For a specific location, use the image to video generator with your own prompt.
Your image is used only to generate your clip and is never used to train models.
Close the deal on video
One image, one click, five seconds. Credits are refunded if a render fails.