Street Light Trend AI — Streetlight Slip Video
A person slips away beside a street lamp, then returns close to the camera with a calm smile. Watch our original five-second example and learn how a portrait becomes a playful night scene.
The idea behind the effect
What is the Street Light trend?
Street Light is a short-video visual joke built around a sudden change in distance. A person starts far from a fixed camera beside a street lamp. They lose their balance and slide out of view; after a brief empty beat, the same person returns as a close-up, looking unexpectedly composed. Streetlight Slip and Street Light Slide describe variations on this format.
The lamp establishes the night setting, while the empty frame creates anticipation. The payoff is the contrast between an exaggerated exit and an unbothered face near the lens. The camera stays still; the person moves between the background and foreground. This is a movement effect, rather than a costume transformation.
What does our original example show?
Our five-second vertical preview features a fictional adult in an olive jacket, cream shirt and dark trousers on an open road at dusk. He leans beside the warm street lamp, loses balance and leaves the left side of the picture. After a short empty interval, he returns close to the camera and smiles. His hairstyle and clothing remain recognizable across the change in distance.
The exit is a broad, comic sideways movement. It is not a precise replay of a social-media clip, a filmed stunt or a realistic fall. The example was newly generated from an original character portrait and includes generated audio. It does not reuse the reference video's performer, footage or soundtrack. The saved video is approximately five seconds, 720p and 9:16.
Why do the pause and return matter?
Without a brief empty interval, the distant slip and final portrait can feel like unrelated shots. Hold the same lamp and background while the person is absent, then bring the same face close to the lens. A small smile or head movement makes the return readable. Extra performers, camera zooms and wardrobe changes can weaken this simple joke.
What photo should I prepare?
Use one sharp, evenly lit portrait of one person, with the face, hair and upper body visible. A front-facing chest-up or waist-up image is a useful starting point. Avoid groups, heavy filters, hidden eyes and a face that occupies only a tiny part of the image. Use a photo you have permission to upload and process.
Your photo supplies the person's appearance, rather than the first frame of the finished video. The night scene is generated separately, in a fixed 9:16 format, so you do not need to take a photo under a lamp or perform a fall. The preview uses an adult human; this effect has not been demonstrated with pets.
A CapCut template associated with the format
A CapCut template titled “ai”, by li88y99, was dated October 21, 2026 and requested one clip in a 9:16 layout when checked on November 1. This is one documented template associated with the format, not proof of a sole original creator. Trends AI Video is independent of CapCut.
An original prompt idea
Place one recognizable person beside a warm street lamp on a quiet night road. Keep the camera fixed. Let the person lose their grip, visibly slide out of the left edge, and leave the frame empty for a short beat. Bring the same person into a chest-up foreground view with a composed smile. Preserve the face, hair and outfit. Request gentle night ambience and a brief sliding sound, without dialogue or a named song. This is a planning example; the website uses a fixed effect prompt and does not offer custom prompt controls.
The Vidu reference-to-video documentation explains the subject-reference and audio controls behind this type of workflow. Reference generation can still change face details, hand contact and motion timing. The preview shows one actual result, rather than a guarantee that every photo will reproduce the same movement.
More ideas to explore
For a different scene, see Tiny Desk AI. For an actual outfit change, explore Halloween Costume Reveal.
From input to output
How to use it
1. Watch the example and check that this effect is marked Available to create. Sign in with a verified email and check your credit balance.
2. Select Street Light in Create video. Upload one clear portrait in PNG, JPEG or WebP format, up to 10 MiB. The output is fixed 9:16; a night photo is unnecessary.
3. Confirm that you have the necessary image rights and consent. Submit once for one approximately five-second, 720p video with generated sound, using 100 credits.
4. Follow the task in Your videos and download the private result when it is ready. If the status is uncertain, check that task or contact support before submitting another request.
Good to know
Best for
Playful portrait clips, a night-scene visual joke, or a short reaction-style reveal. Use the result as an AI interpretation, rather than evidence that a real person performed a stunt.
Limits & considerations
AI can change facial details, hands, clothing and the exact exit or return timing. The slip may look exaggerated, and a brief pose or anatomy artifact can appear. The scene is generated from the portrait, not a frame-accurate edit of a source video. Audio is generated; no named social-media song, dialogue or accurate lip sync is promised. Use permitted images and do not imitate the slip physically. Check current availability before purchasing credits. The complete result costs 100 credits; uncertain or failed tasks may need support review before credit handling is confirmed.
Questions about this effect
Is Streetlight Slip a clothing transformation?
The defining change is distance: a person leaves the lamp-side background and returns as a close portrait. Keeping the same outfit helps connect the two views.
Do I need a night photo or a video of myself falling?
No. Upload one clear portrait. The generator creates the lamp and night setting around the referenced person, with a fixed vertical 9:16 output.
What does the original preview include?
A fictional adult in an olive jacket, a comic exit beside a lamp, a short empty beat and a smiling close-up return. It is approximately five seconds at 720p with generated audio.
Can I use a pet or several people?
This template is designed for one human portrait. Its verified example uses an adult person; pet and multi-person versions have not been demonstrated.
Will my face and the motion match exactly?
No exact match is guaranteed. Face details, balance, hand contact, movement and the pause can vary between generated results.
Is the original trend music included?
The example has generated audio. The workflow requests night ambience and movement effects, without a named track or dialogue. You can add music separately only if you have the relevant rights.
How much does one video cost?
When the effect is available, one complete private video uses 100 credits. The Starter pack is US$9.90 for 100 credits, paid once through PayPal with no automatic renewal. Check current availability before buying.