Orelon logoOrelon
价格

AI Shot Lists for Wedding Cinematography: A Practical Guide

2026年11月14日 · 作者:Orelon Team

探索 AI 视频模板

浏览社区创作获取灵感,打开任意模板即可在 Orelon 中继续创作。

Build reliable wedding shot lists with AI assistance: a five-block coverage framework, prompt techniques for consistent looks, pre-visualization tests, and common mistakes.

A wedding film is won before the first frame is exposed. It is won by the shot list: the quiet document that tells you where to stand, what to protect, and when to stop planning and start watching. The best wedding cinematography is not the most elaborate coverage — it is the most deliberate. AI has made that deliberation cheaper, because a model can draft a full day of setups, cross-check them against a venue and a schedule, and produce visual references in the time it used to take to sketch four frames on a napkin.

But a generated list is not a plan. It is raw material. This guide is about turning that material into something a real crew can shoot, in real light, with real guests stepping into your frame.

Start With the Coverage Problem, Not the Camera

Every wedding has a small number of irreplaceable moments: the first look, the walk down the aisle, the vow exchange, the first touch after the pronouncement, the first dance, the exit. Everything else is supporting material. A shot list exists to guarantee those moments are covered from at least two angles while leaving enough slack for the unexpected.

Before you ask a model for anything, write down three things:

  1. The non-negotiables. Five to eight moments the couple has named, plus the ones they will care about later but did not name — grandparents, a sibling's toast, the family recipe table.
  2. The constraints. Ceremony length, venue light at each hour, whether you can move during vows, whether there is a second shooter, how far the portrait location sits from the reception.
  3. The visual reference. Three to five films or stills that describe the tone. Not "cinematic" as a word — cinematic as a set of decisions about contrast, lens length, camera height, and pace.

That brief is what makes an AI-assisted list useful. Without it, you get a generic wedding montage checklist that looks like every other wedding montage checklist. With it, you get a plan that reads like it was written by someone who has already been to that venue and knows where the light falls at four in the afternoon. Write the brief in plain language, but be specific about place and time; those two details do more for output quality than any list of adjectives.

What an AI-Assisted Shot List Actually Does

A generative model is good at three tasks in pre-production, and bad at a fourth.

It is good at breadth: listing forty coverage options for a ceremony in a stone church with a north-facing altar, including the ones you forget when you are nervous.

It is good at translation: turning a mood reference into concrete descriptions — "low-angle handheld, 35mm equivalent, backlit haze from the doorway, shallow focus on the veil."

It is good at sequencing: proposing an order of setups that respects the timeline instead of the other way around.

It is bad at judgment on the day. It does not know that the grandmother is frail, that the officiant stands in front of the best window, or that the couple will want thirty seconds alone after the ceremony. Those come from you.

The practical division of labor: let the model produce volume and structure, then edit ruthlessly. A good list for a full day usually ends up around thirty-five to fifty setups after you cut a generated list in half.

The Five-Block Framework for a Wedding Day Shot List

Most wedding days break into five coverage blocks. Treat them as separate lists with separate goals, because the camera behavior that works in one block fails in the next.

Block 1: Prep and Getting Ready

Goal: texture and anticipation. The common mistake is shooting too much of the room and too little of the hands.

  • Details in context, not on a table: rings on the invitation, the dress hanging in a doorway with morning light behind it, shoes next to a window.
  • Hands: buttoning a cuff, fastening an earring, a mother adjusting a veil.
  • Faces before the reveal: someone looking out of a window, the quiet moment of a breath.
  • One wide that establishes the space for later editing.
  • Movement: a slow push through a doorway, or a handheld follow behind the subject.

Ask the model for "detail shots that show anticipation rather than objects" and you will get better results than asking for "getting ready shots."

Block 2: Ceremony Coverage

Goal: complete coverage with no visible camera movement during the vows. List this block as a matrix: position, lens, what happens, and what could go wrong.

  • Position A: back of the aisle, wide, locked off on a tripod. This is your insurance shot.
  • Position B: side aisle, medium long, capturing the couple's faces and the officiant's profile.
  • Position C: a reaction angle on the front row, shot sparingly and quietly.
  • Insert list: hands, rings, the program, the aisle runner, the flower girl's hesitation.
  • Sound note: cues for the officiant's microphone and the reader's position, because audio failure ruins more ceremony films than bad framing.

Block 3: Portraits and Couple Time

Goal: a small number of frames that feel like the couple, not like a pose. Ten strong frames beat forty variations.

  • One wide establishing the landscape or architecture.
  • One medium with both faces readable and the background still doing work.
  • One close frame at a time, alternating between the two people.
  • One movement shot: walking, turning, or a slow drift past them.
  • One frame that includes the venue in a way that will matter in ten years.

Prompt the model with the specific location and time of day, and ask for shot descriptions that note where the light comes from. "Golden hour on a west-facing terrace" produces usable ideas; "romantic portraits" produces nothing.

Block 4: Reception Energy

Goal: the room as it felt, not a highlight reel of reactions. Vary distance and height deliberately.

  • Wide from a high corner for scale during the entrance and the first dance.
  • Low, wide, close to the floor for dancing, so the crowd reads as a crowd.
  • Detail inserts: place cards, the cake before it is cut, hands on a glass during a toast.
  • Reaction coverage: two or three faces during each toast, chosen in advance by seat position.
  • One slow-shutter or movement frame per hour to give the edit breathing room.

Block 5: Golden Hour, Night, and Exit

Goal: an ending. This block is the one most often improvised and most often regretted.

  • A ten-minute golden hour window scheduled in advance, with a backup location if the light fails.
  • One night exterior using practical lights as the key.
  • A final quiet frame of the two of them, away from the crowd, before the exit.
  • The exit itself: one wide, one from behind, one from the front if the space allows.

Writing Prompts That Keep the Look Consistent

Consistency across a wedding film comes from repeating a small set of decisions. Write them once as a look block and paste it into every prompt so the model never drifts.

A look block has five lines:

  • Format and lens feel: "35mm equivalent, slight halation, shallow depth at f/2."
  • Palette: "warm neutrals, muted greens, skin tones kept natural, no teal-orange push."
  • Light: "available light with one soft key; strong backlight and haze when the sun is low."
  • Camera behavior: "handheld with intent, slow push-ins, no whip pans, no drone spirals."
  • Texture: "fine grain, soft contrast, realistic motion blur."

Then write shot prompts as sentences with a subject, an action, a framing, and a light direction. "Bride's mother fastens the veil, medium close-up from behind, north window light rimming her hands, shallow focus" will always beat "beautiful veil shot." If you want a starting library of prompt structures, the prompt library is a reasonable place to borrow phrasing.

Two rules keep prompts from flattening the day. First, never describe a moment you have not confirmed will happen. Second, always include the constraint — time of day, indoor or outdoor, direction of light — because constraints are what make a generated frame plausible.

Iterate in one direction at a time. If the framing is right but the skin tones are wrong, change only the palette line. If the light is right but the motion feels theatrical, change only the camera behavior line. Prompts that change three variables at once teach you nothing, and you will not be able to reproduce the result on a different venue next month.

Pre-Visualization: Turning the List Into Motion Tests

Once the list exists, the next step is testing the look before the wedding, not during it. Generate short motion clips for the five or six setups you are least sure about: the backlit doorway, the ceremony from the side aisle, the night exit.

Use these tests for three decisions:

  • Lens and distance. Does the wide establish the space, or just make everyone small?
  • Pacing. Does a slow push feel right, or does it feel like a drone shot with extra steps?
  • Color. Do skin tones survive the venue's mixed lighting?

Generate stills first with the AI image generator to lock framing, then animate the two or three that matter with the AI video generator. Keep clips short — three to five seconds is enough to judge a look. Treat them as pre-visualization, never as inserts in the final film.

The Coverage Plan Versus the Moment

The most important line in any shot list is the one that says "put the list away."

A plan earns its value by making the day predictable enough that you can notice what is not on it: the father of the bride standing alone in the hallway, the ring bearer asleep on a chair, the couple's hands shaking slightly during the vows. Those frames are never generated, because they are not patterns.

A workable rule: shoot the plan for the first ten minutes of each block, then follow attention rather than the document. Keep a running note of what you missed so it does not become a pattern across the season. If you notice the same gap three weddings in a row — say, no usable frames of the couple's friends — add it to the brief rather than blaming the day.

A Two-Week Working Timeline

Ten days out. Collect the couple's priorities, the venue's floor plan, the ceremony order, and the light at each hour. Write the brief.

Eight days out. Generate a broad list. Cut it to fit the actual timeline. Assign positions to a second shooter if there is one.

Six days out. Generate reference stills for the uncertain setups. Adjust the look block until the references match the couple's taste.

Three days out. Walk the venue if possible. Mark positions on the floor plan. Confirm the portrait window against sunset.

Day before. Print a one-page version: only the non-negotiables and the position map. Everything else stays in your head or on your phone.

Day after. Review what you shot against what you planned. Add one line to the template about what you would change.

This cadence keeps pre-production from expanding to fill all available time, which is the most common failure mode once generative tools enter the workflow.

Mistakes That Flatten Wedding Films

  • Generating lists without constraints. A model with no venue, time, or light information produces a montage of generic setups.
  • Over-listing. Sixty setups you cannot shoot becomes thirty setups you shoot badly.
  • Confusing pre-visualization with the deliverable. Generated clips are a thinking tool. Using them as coverage is a different film with a different promise.
  • Ignoring audio. The best-looking ceremony footage is unusable if the vows are not clean. Put microphone cues in the list.
  • Shooting the list instead of the room. If a plan keeps you from turning around when something real happens, the plan is wrong.
  • No repetition. A film feels authored when three or four visual motifs recur — the doorway wide, the hands insert, the slow push. Decide on the motifs and schedule them.

Choosing Tools Without Over-Tooling

You need three things: a place to generate references, a place to organize the plan, and a way to keep the look consistent across both.

For references, a browser-based generator with image-to-video and style controls covers almost every pre-production need. Browse video templates if you want a structural starting point rather than a blank prompt box. For organization, a plain document with five blocks and a position map beats most production apps, because it is open on the day and works without a battery.

If you compare platforms, judge them on three questions: can it hold a consistent look across multiple generations, can it animate a still you already framed, and can it output short clips fast enough to be useful before the wedding? Everything else is decoration. The alternatives overview helps calibrate expectations, but the real test is running your own look block through two or three tools and comparing skin tones side by side.

FAQ

Do AI shot lists replace a cinematographer's judgment? No. They replace the blank page. Judgment — what to prioritize, when to abandon the plan, how to read a room — remains entirely human.

How many setups should a wedding day shot list contain? Roughly thirty-five to fifty for a full day with one shooter, weighted toward the ceremony and the portrait window. The list should fit the timeline, not the other way around.

Can generated clips be used in the final wedding film? They can work as title backgrounds or transitions if the couple knows and approves. As a substitute for coverage, no — the film's value is that it documents what actually happened.

What is the best way to keep visuals consistent across setups? Write a five-line look block covering lens feel, palette, light, camera behavior, and texture, and reuse it in every prompt. Consistency is a repetition problem, not a model problem.

How far in advance should pre-visualization happen? A week out is enough for most weddings. Earlier is better if the venue has unusual light, because it gives you time to move the portrait window.

What if the couple wants a trend-heavy look that will date quickly? Show them two reference sets: one trend-forward, one built on skin tones and natural light. Most couples choose the second once they see both, and the film ages better.

Build the List, Then Forget It

The point of an AI-assisted shot list is not to make the wedding more controlled. It is to make the day predictable enough that you can be surprised by it. Generate broadly, cut hard, test the setups you are unsure about, then put the document away and watch the room.

Orelon is built for that pre-production stage: cinematic ideas in motion, from a single reference still to a short motion test you can judge before the wedding day. Start with the AI video generator, bring your look block, and see whether the frames feel like the film you described to the couple.