How to make a premium event recap video with gold branding

Last updated 2026-08-24 · 7 min read

What is the short version?

A recap video for a course or a ceremony is selling next year's intake on the strength of last year's photographs. The material is fixed, so everything that makes it feel premium comes from the treatment: the grade, the gold, and typography that is actually beautiful rather than merely present.

1. Go to Vizard Agent and upload the photos and footage you have.
2. Say the look — the palette, the register, the feeling.
3. Give it the logo and the wording for the titles.

What do you need before you start?

Last year's material and a description of the register. Vizard Agent generates the backgrounds, treats the logo and grades your photos, so a modest set of images is enough. What matters most is stating the tone precisely, because luxury and warmth are achieved through restraint rather than through effects.

• The photos and footage. From the event and the ceremony.
• The palette. Dark navy and gold, or whatever your identity is.
• The logo. Vizard Agent will cut it out and treat it in gold.
• The title wording. In whichever language and script.
• The length. Under thirty seconds for a recap.

What do you type into Vizard Agent?

Name the palette and the feeling together. Vizard Agent uses that pair to drive the generated background, the grade on your photographs, the transition flashes and the ornament on the title cards, so a phrase like "luxury, warm, spiritual, dark navy with gold" governs the whole build.

Prompt

Variants worth knowing:

• A generated background. For the title cards, matched to the palette.
• An acapella or instrumental score. Some events need music without instruments.
• Gold ornament on the cards. Drawn rather than borrowed from a template.

What does Vizard Agent actually do?

Here is the order Vizard Agent worked in on a real event recap. An unusually large part of the job was typography, because the titles are in a script where the letters have to join correctly and most fonts get that wrong.

1. Downloads and inspects every attached photo and reviews them as a grid.
2. Generates the music, the sound effects and the cinematic background in parallel.
3. Cuts out the logo and refines its edges and transparency into a clean gold mark.
4. Analyses the music to confirm it suits the register of the event.
5. Downloads three premium typefaces for the script and installs the text-shaping libraries.
6. Tests each font's letter joining on rendered frames, comparing them with and without diacritics.
7. Designs the gold title cards for every scene, then enlarges and sharpens them after review.
8. Draws the gold ornament and adjusts the titles precisely, checking each scene on frame.
9. Grades the real footage to the warm cinematic palette, generates the transition flashes, then re-crops every photo to favour the faces and moves the titles to the lower third so nobody is covered, and re-renders the whole film.

Step 9's second half is the fix that matters most to the people in the video. A title card centred beautifully in the frame will sit across somebody's face, and re-cropping to favour faces and dropping the text lower costs one pass and changes how the film feels to everyone who attended.

What does the result look like?

From the run this page is written from, probed on the delivered file: 1920x1080, H.264, 30fps, 28.0 seconds, AAC audio. Widescreen, twenty-eight seconds, last year's photographs graded to a warm palette against a generated background, with gold title cards, drawn ornament and a closing invitation.

Twenty-eight seconds is short for a recap and correct for a promotion. The video exists to make next year's course feel worth joining, and a longer edit turns it into a record of an event that has already happened.

When does this not work well?

This format is built entirely from photographs of people at an event, which makes consent the first question rather than a technical footnote. Vizard Agent treats the material beautifully, and how it may be used is settled before any of that matters.

• Attendees appear by name and face. Their agreement to appear in promotion is a separate thing from attending.
• Religious and cultural material needs review. Get the wording checked by someone inside the community.
• Scripts with joined letters break in most fonts. Vizard Agent tests them, and unusual scripts remain the risk.
• Photographs limit the pacing. Stills with motion applied are not footage, and a long edit exposes that.
• A grade cannot rescue a bad photo. The warm palette flatters; it does not fix.

How do you fix a result that came back wrong?

Name the card or the photo. Vizard Agent keeps the treated logo, the tested fonts, every title card version, the graded photographs and the crop adjustments, so a wording change or a re-crop is a re-render rather than a rebuild from the source images.

• "The title covers her face." Re-cropped and the text moved lower, as this run did throughout.
• "The letters are not joining." Another typeface, already downloaded and tested on frame.
• "The gold is too bright." Re-treated against the palette.

How does Vizard Agent compare to doing it yourself?

By hand this means finding a font that handles the script, discovering it does not join letters correctly, trying two more, and then building each gold title card individually in a design tool. Meanwhile the crop on every photograph is left as-shot, which is how titles end up sitting across the faces of the people you are trying to celebrate.

Common questions

Can it work from photographs alone? Yes. Vizard Agent applies motion and grades them, and a few clips add variety if you have them.

Will the titles be in my language? Yes, and Vizard Agent tests the typeface's letter joining on real frames before committing to it.

Can it treat our logo in gold? Yes. It cuts the mark out cleanly and refines the edges before applying the treatment.

How long should a recap be? Under thirty seconds when the goal is filling next year's intake.

Why does the typography take so long? Because in scripts where letters join, most fonts render them separately or drop the diacritics, and it looks wrong to every native reader. Vizard Agent downloads several and checks each one on a rendered frame rather than trusting the font's name.

Will it cover people's faces? No. Vizard Agent re-crops to favour faces and places the titles clear of them.

Can it generate a background? Yes. Vizard Agent generates one matched to the palette you name for the title cards to sit on.

Does Vizard Agent check the finished film? Yes. It reviews the scenes across the whole timeline and re-rendered this one after adjusting every crop.