00:00 / 00:00

Adding a New Element

Lesson 410 lessons

I didn't want the lizards just stuck behind me — I wanted a *reveal. *So I asked for a zoom-out first: tight on them climbing first, then pulling back to show me fully with the right lip-sync. prompt snake >>: Original clip — young man with curly dark hair and a moustache, rings on his fingers, wearing an open light grey pinstriped short-sleeve shirt over a black tee, standing in a sunlit rooftop parking lot in Almaty, gesturing and talking directly to camera; behind him the beige Hotel Kazakhstan tower with its crown-shaped top, a Soviet apartment block at left, parked cars and green trees, low handheld angle with a slow push-in, warm low-sun daylight. Preserve the man's identity, face, moustache, wardrobe, rings, performance and exact lip-sync, plus the framing, lens character and handheld motion of the original take exactly. Change only the background behind him: add many large reptiles climbing the hotel tower, and prepend a 1-second telephoto opening that snaps out to the original composition. >>: Reference of a real large monitor lizard — pebbled scaly skin, long claws, heavy tail, true reptile anatomy. Appearance, scale-texture and color reference only; ignore the photo's background and lighting, do not use it for the environment. Photoreal. 16:9. 8s. Warm low-sun evening daylight, soft long shadows, gentle film grain — grade matched to the source plate. NON-IP — generic large monitor lizards, not based on any brand or character. SFX and source dialogue only. One continuous shot, no cuts. For the first 1 second hold on a long telephoto lens (~300mm look, compressed perspective) framed tight on the upper facade of the beige Hotel Kazakhstan tower: dozens of large photoreal monitor lizards, each roughly car-sized, swarming up the building en masse — claws hooking window ledges and concrete, heavy tails dragging, pebbled scales catching the low sun, bodies overlapping as they crawl higher, real soft-edged contact shadows on the facade. At the 1-second mark the camera snaps a hard, fast zoom-out / quick pull-back, decompressing perspective and craning down to land exactly on the source composition — a 100% match of the original framing: the low handheld angle with the man in the foreground, tower behind, same headroom and horizon. From 1 second onward play the man's preserved take: he speaks directly to camera, lips matching the source exactly, saying clearly: "I can put something behind me that definitely shouldn't be there. Insane right?" — keep his performance, gestures, timing and lip-sync precisely as in the source, smiling and completely unfazed and oblivious to the creatures, while the source's own slow handheld push-in continues. Behind him the lizards keep steadily climbing the tower for the rest of the shot, more of them crawling into frame on the facade. Keep the lizards integrated into the plate: same warm key from screen-right and same color temperature as the man, real contact shadows where claws grip, matching atmospheric haze and depth of field so they sit at true distance, fully photoreal real scale detail and true reptile anatomy, never CG, plastic or cartoonish. Lock-down: the man's face, identity, moustache, wardrobe, rings, expression, gestures and lip-sync are unchanged; from 1s onward the framing, lens and handheld push-in are a 100% exact match of the source — only the lizards on the tower are added. SFX and source dialogue only: open with a quick burst of distant claw scrapes and scuffs on concrete and glass and scattered low reptilian hisses over faint city ambience; at the 1s zoom-out the man's preserved voice comes in and stays dominant — "I can put something behind me that definitely shouldn't be there. Insane right?" — the scrapes and hisses continuing softly underneath, a low structural rumble far in the background. Let’s see what we got. That's how you add something into your shot — plus a camera move you never filmed. Now let’s move to the one you've been waiting for, and the hardest thing we'll do all video: the handheld cinematic showcase. The kind of big-budget CGI shots that normally need a whole studio. This is the hardest one so far. On a locked-off or steady shots we have already seen, the AI has it easy — the frame barely changes, so it just paints the effect in. But the second the camera's in your hand, everything's moving chaotically — the angle, the parallax, the shake — and the effect has to track all of it without falling apart. That's this stage: fully handheld, me and the camera moving the whole time. And here's the whole trick: I shoot each clip already knowing what I want it to see in it, then one simple prompt rebuilds the entire world around me. Same clip in, a different planet out. Let’s see the first one **Wing-walker on a biplane** I start simple — change my wardrobe and drop me somewhere that couldn't exist: a wing-walker on a flying biplane over floating mountains. I have this one prompt prompt 1 wing walker
Recreate

Adding a New Element

Higgsfield Academy

Caption

Swapping the world around you is one thing — now try dropping a brand-new person into the shot. The workflow is the same tool as before, Edit Video, just pointed at a different kind of change.

This time the ask is to add an older version of himself walking by in the background of the clip. To make sure the AI gets the likeness right, he attaches a character sheet — a reference image of what that older version of him should look like — alongside the source footage. With the reference in place, the prompt itself is short: "add an old man." Hit Generate, and the passerby shows up in the shot, folded into the scene as if he'd been filmed there.

The core idea carries over from the earlier background swaps: everything else in the plate — his own performance, the framing, the lighting — stays locked, and only the new element is introduced. The difference is that instead of replacing the environment, this pass adds something to it, which is the same technique to reach for any time you want to insert a person into footage you already shot rather than rebuild the whole scene around it.

Previous Next