How to Convert Horizontal Video to Vertical (16:9 to 9:16)

To convert horizontal video to vertical for Shorts, put it on a 1080x1920 (9:16) frame and move the crop with the speaker: by hand, with keyframes or with AI.

Updated: 29 October 2026

We make LumiClip, so weigh our view accordingly; we've noted where an editor or another tool is the better pick.

This guide compares the four ways to turn 16:9 footage into 9:16: a static crop, a blurred-background fit, editor keyframing and AI speaker tracking. It also shows the framing mistakes that ruin two-person shots, and the export settings YouTube Shorts expects. Platform rules were checked on official help pages on 29 October 2026.

What is the best way to convert horizontal video to vertical for Shorts?

For talking videos, use a reframer that follows the active speaker; for a single locked-off shot, a static center crop is enough. Set a 9:16 frame at 1080x1920, move the crop so the person talking stays in frame, keep captions inside the safe area and export under YouTube's 3-minute Shorts limit. Premiere Pro, DaVinci Resolve Studio, CapCut and LumiClip can all do the reframing automatically.

What happens to a 16:9 video when you crop it to 9:16?

A full-height 9:16 crop keeps less than a third of a 16:9 frame, so you have to decide what to keep in every shot. In a 1920x1080 video, a vertical slice the full height of the frame is only about 608 pixels wide. Everything outside that slice is gone.

That has three practical effects:

• Something always gets cut. A wide shot of a presenter and their slides, or two people at a table, can't fit in one vertical crop at full size.
• 1080p sources lose sharpness. That 608x1080 slice has to be scaled up about 1.8 times to fill a 1080x1920 frame, so faces look softer.
• 4K sources survive the crop. From a 3840x2160 video, the same slice is about 1,215x2,160 pixels, which is more than a 1080x1920 export needs. If you know a vertical version is coming, record in 4K.

Which 16:9 to 9:16 conversion method should you use?

Pick the method by how much the subject moves and how many people are on screen. A static crop suits one person who stays still. Keyframing suits short, important clips you want to control. AI speaker tracking suits conversations and anything you convert every week. Here is how the four options compare, with the same labels for each.

Method 1: Static center crop

• How it works: set a 9:16 canvas, scale the video to fill the height and leave the crop in the middle. Most editors and phone apps do this in one click.
• Effort: under a minute per clip.
• Best for: a single talking head who stays centered, such as a webcam video or a locked-off camera.
• Watch out for: anyone who moves, gestures off-center or shares the frame with someone else. The crop doesn't follow them.

Method 2: Fit with a blurred background

• How it works: keep the whole 16:9 frame in the middle of the vertical canvas, then fill the empty space above and below with an enlarged, blurred copy of the same video.
• Effort: a few minutes, and reusable as a template.
• Best for: footage where the full width matters, such as a screen recording, a slide or a wide sports play.
• Watch out for: the picture only fills about a third of the phone screen, so faces are small. It reads as a converted video, not a native vertical one.

Method 3: Keyframing in a video editor

• How it works: put the clip on a 9:16 timeline and add position keyframes so the crop moves with the subject. Premiere Pro's Auto Reframe generates those keyframes for you, with motion presets from Slower Motion (talking heads) to Faster Motion (sports). DaVinci Resolve has Smart Reframe, which is a Studio-only feature, and CapCut has Auto reframe.
• Effort: low with auto reframe, high if you keyframe every speaker change by hand.
• Best for: editors who already work in Premiere or Resolve, and one-off clips where you want frame-by-frame control.
• Watch out for: editor auto reframe follows action, not audio. In a two-person shot it may track whoever moves, not whoever talks, so you still check every speaker change.

Method 4: AI speaker tracking in a clipper

• How it works: a clipper analyzes who is speaking and plans a 9:16 crop path that follows them. LumiClip, an AI video clipper that turns long videos into short vertical clips for TikTok, Reels and YouTube Shorts, combines audio speaker cues with visual tracking, per its active-speaker reframing page.
• Effort: low. You review the result instead of building it.
• Best for: podcasts, interviews, webinars and anything you convert in batches, because it also picks the clips and adds captions.
• Watch out for: slides, products and unusual camera angles may still need a manual crop adjustment. If you need precise, artistic framing on one hero clip, an editor gives you more control.

How do you convert horizontal video to vertical in a video editor?

Make a 1080x1920 timeline, reframe every shot so the subject stays inside the tall frame, move titles and captions into the safe area, then export. These steps work the same way in Premiere Pro, DaVinci Resolve and CapCut. Only the menu names change.

Step 1: Finish the horizontal edit first

Lock the 16:9 cut before you reframe. Every change you make to the edit after reframing means reframing that section again.

Step 2: Create a 9:16 copy of the timeline

Duplicate the sequence and set the frame size to 1080x1920. In Premiere Pro, Sequence > Auto Reframe Sequence creates a reframed duplicate for you. In Resolve, set the timeline to a custom vertical resolution with mismatched files set to Scale full frame with crop, so there are no black bars.

Step 3: Reframe each shot separately

Give every shot its own framing. Never let a crop slide from one shot into the next across a cut, because it looks like a mistake on screen. If the subject doesn't move, leave the crop still.

Step 4: Move graphics and captions

Lower thirds and titles designed for 16:9 will hang off the edge of the vertical frame. Lay them out again inside the center of the frame, clear of the platform buttons on the right and the caption area at the bottom.

Step 5: Trim and export

Vertical versions are usually shorter. Cut anything that only worked in wide shot, then export an MP4 at 1080x1920.

How do you reframe a video to vertical with AI in LumiClip?

Upload the 16:9 video, let LumiClip pick the moments and plan the vertical crop, review each speaker change, then export or schedule. The AI 9:16 reframing tool handles the crop path, captions and safe zones in one pass, so the steps are mostly review.

Step 1: Add the horizontal video

Upload the highest-quality file you have, or paste a YouTube link. The free plan accepts uploads up to 1 hour, and Starter and Pro accept up to 4 hours, per the LumiClip pricing page.

Step 2: Let the AI find clips and plan the crop

LumiClip scores moments on hook strength, speech clarity, visual variety and emotional peaks, then reframes each clip. According to its reframing page, it uses single-speaker framing when one person talks, split screen only when voices genuinely overlap, picture-in-picture when someone reacts silently, and subject tracking when no one is speaking.

Step 3: Review every transition

Play each clip and watch the moments where the speaker changes. Check that the crop lands on the right face, that lower thirds and on-screen text aren't sliced, and that split screen only appears on real back-and-forth. Adjust the crop in the editor where needed.

Step 4: Check captions and branding

Captions are word-level and animated, in 30+ languages with automatic detection. Fix names and jargon in the transcript, then apply a Brand Kit preset so every clip uses the same fonts, colors and logo.

Step 5: Export or schedule

Download the vertical MP4s, or connect your accounts and schedule clips to YouTube Shorts, TikTok, Instagram Reels and LinkedIn. Per the reframing page, the same pass can also export 1:1 and 4:5 versions for feed posts.

What framing mistakes ruin two-person shots?

The most common mistake is a center crop that lands on the empty space between two people. Wide two-shots are the hardest footage to convert, because a 9:16 frame can only show one face at full size. Watch for these:

• The empty middle. In a wide shot of two people at a table, the center is the gap, a microphone or a plant. Neither person is in frame.
• Following movement, not speech. Tracking that follows motion jumps to whoever gestures, even when the other person is talking.
• Whip pans between speakers. A crop that slides fast from one face to the other on every reply is tiring to watch. Cut between them instead.
• Split screen for the whole clip. Stacking both people when only one talks halves the speaker's size for no reason. Save it for genuine back-and-forth.
• Cutting out the reaction. If the listener's laugh is the point of the moment, use picture-in-picture or cut to them.
• Captions over faces. In split layouts, move captions to the middle band, between the two frames.

If your footage is a video podcast, our guide to turning a two-speaker podcast into YouTube Shorts covers recording setups, layouts and linking each Short to the full episode.

What size and length should a vertical Short be?

Export at 1080x1920 (9:16) and keep the clip at 3 minutes or less. Per YouTube Help, a Short must be up to 3 minutes long with a square or vertical aspect ratio. A 16:9 file doesn't meet that rule, which is why you convert it first.

The same 1080x1920 file works for TikTok and Instagram Reels, so you only need one vertical export per clip.

FAQ

Can I upload a horizontal video as a YouTube Short?

Not as a 16:9 file. YouTube says a Short needs a square or vertical aspect ratio, so convert the video to 9:16 (or 1:1) before uploading.

Does converting 16:9 to 9:16 lower the quality?

It can. A vertical crop of a 1080p video is scaled up about 1.8 times, so it looks softer. Cropping from a 4K source avoids that.

Is there a free 16:9 to 9:16 converter?

Yes. CapCut offers Auto reframe, and LumiClip's free plan includes 60 credits once (1 hour of video) with 720p, watermarked exports. Starter removes the watermark and adds 1080p exports and scheduling.

Can AI reframe 4:3 or ultra-wide video to vertical?

LumiClip's reframing page says it handles 4:3 and ultra-wide sources as well as standard 16:9, and reframes anything wider than 9:16.

How do I keep both people in frame when converting to vertical?

You usually can't at full size. Use split screen when both talk at once, and follow the active speaker the rest of the time.

Which method should you choose?

Use a static crop for solo, centered footage, an editor for hand-crafted one-offs, and AI speaker tracking for conversations and weekly batches. Whatever you choose, check every speaker change, keep text inside the safe area and record in 4K when you can.

If you convert long videos to Shorts every week, try LumiClip free on one video, then compare plans on the pricing page. Starter is €9.99/month for 360 credits (6 hours of video), as of October 2026; the pricing page showed euro prices when we checked from Europe. To skip the uploading step entirely, see how to auto-clip a YouTube channel to TikTok, or compare watermark-free plans in our cheapest AI video clipper price check.

Cover photo by Thiago Zanutigh on Unsplash