Street Light Slide AI Video Generator

How do I make the street light slide AI video using just one photo of myself?

Choose a portrait that shows how you want to look at the end of the clip. The street light slide AI video generator supplies the empty path, the lamp and the motion, so a daylight photo works. The CapCut template titled “ai,” published by li88y99 on October 21, lists one clip in a vertical frame. If your editor asks for video, you can use the original still here. Open the photo picker, select a model and resolution, check the credit quote, then generate. You can reopen the finished MP4 from Creations.

1. Pick one portraitShow the whole hairline and both shoulders of the consenting adult who will appear under the lamp.
2. Choose the exportMini at 480p costs 10 credits. Choose 720p for more detail in the final close-up, then confirm permission.
3. Play it with soundCheck the pause and the final face. Download the five-second MP4 from the result or reopen it in Creations.

What is that street light slide AI trend everyone's doing on TikTok?

Watch the lamp rather than the face for the first second. It stays in place while a small figure skids left along the pavement. For a moment the path is empty. Around a second and a half in, the subject enters from the lower left and comes to rest close enough to see their expression. A widely shared October 22, 2026 clip paired this switch in scale with the excerpt labeled “original sound - naddy.” Some street light trend videos send the camera down the road. In these three takes, the person makes the move.

1. 
   0.04 sSlide out
2. 
   1.00 sHold on the path
3. 
   2.25 sSettle under the lamp

How much does it usually cost to make one street light slide AI video?

One five-second take costs 10 credits on Seedance 2.0 Mini at 480p, or 22 credits on the same model at 720p. Both include the motion guide and the finished soundtrack. On the Studio monthly plan, 10 credits represent about $0.28 of the $49.90 monthly allowance. That figure describes usage on that plan. Starting from an empty balance, the $5 trial pack buys 24 credits, enough for two Mini 480p takes or one Mini 720p take. Check the selected row before you generate a different version.

What's the best AI tool for making the street light slide trend from a photo?

If you want the original-looking slide, play a sample and watch what happens to the streetlamp. In the fixed-camera version you can follow the pole through the wide shot and the close-up. For a single selfie, the next decisions are the length of the hold, the export quality and what you have to buy before rendering. MakeViral lists one five-second 720p setting. Videosi publishes a six-second Standard example price and several output shapes. TryH3 keeps the vertical five-second cut across its model choices. Each service sells credits on its own scale.

Why does my face look different in the street light slide AI video?

In the wide shot your face occupies only a few pixels. Judge the likeness when the subject has stopped under the lamp, about two seconds in. A portrait with hair across one eye, a strong beauty filter or a hand against the cheek leaves the model to guess details that become obvious in close-up. Use even lighting and leave room around the jaw. Higher resolution can sharpen the final portrait, but it cannot recover a face hidden in the input. This scene takes one person. It cannot stage two friends together or guarantee an identical expression.

Can you give me a copy-and-paste prompt for the street light slide AI trend?

The motion cue that matters most is the moment the pavement is empty. If you only ask someone to slide toward the camera, a video model may turn the whole trip into a walk. Give the distant exit and the close arrival separate times. Copy the first prompt to see the instructions TryH3 pairs with your portrait, the empty path and the motion guide. If you use a photo-capable Kling or Veo workflow without those references, try the second prompt. The timing will be less predictable.

Make one five-second vertical shot. @Image1 is the only performer. Keep this person's face, skin tone, hair length and texture, build and outfit in both appearances. @Image2 supplies the empty night path, pavement, grass, tree and cool overhead streetlamp. Render that setting in natural colour. Use @Video1 as a depth and timing guide. Keep the camera locked to the same position, focal length and horizon.

0.00–0.40 s: The performer starts very small beside the lamp base in the distance. Low to the ground, they slide sideways out through the left edge. Their feet do not take steps toward the camera.
0.40–1.45 s: Hold on the empty path. Keep the tree, curb and lamp fixed. The subject is completely out of view.
1.45–2.10 s: The same person glides into the foreground from the lower left, rising naturally into a chest-up close-up under the lamp. Follow the guide's speed and direction. Let loose hair and clothing settle with the movement.
2.10–5.00 s: Hold the close-up. The performer looks straight into the lens with a relaxed mouth and a small settling sway. Preserve the features and clothing from @Image1 in the lamp light.

Keep the frame continuous and the street in colour. Exclude camera pushes, morphs, duplicate bodies, slow walking, depth shading, lettering and overlays.

Make a five-second vertical video with the adult in the supplied portrait. Preserve their face, hair and outfit. Lock the camera beside a paved path at night. A cool white streetlamp and a slender tree stand left of center, with grass on the right. In the first half second a tiny version of the subject skids low out of the distant left edge. Leave the path empty until about 1.5 seconds. Bring the same person in quickly from the lower left, now close to the lens. By 2.1 seconds they settle into a chest-up portrait. Hold their gaze with a slight sway until the fifth second. Keep the background still and the body moving sideways. No walking toward the lens, camera zoom, text or extra people.

What's the best photo to upload for the street light slide AI effect?

A chest-up portrait in soft, even light gives the model both the face and the neckline that will fill the last three seconds. A full-body photo can work if you can still see the eyes when you zoom in. Leave a little space above the hair and choose one person with no phone in front of their face. The photo background can be indoors because the path comes from a separate image. If two of you want matching clips, start with two individual portraits and make one take per person.