Video Remastering

SDR2HDR API — convert standard video to HDR at scale

Restore brightness, widen color gamut and recover lost detail from SDR content with an AI model trained on professionally graded masters. No manual color work required.

VideoRemastering

Before & after

See the difference

Drag the handle to compare the original SDR frame with the tone-mapped HDR output.

HDR output
SDR input
SDR inputHDR output

What it does

SDR2HDR analyzes every scene and rebuilds it in the high dynamic range domain: adaptive tone mapping lifts dark and clipped regions into a natural HDR range, a learned wide-gamut path expands Rec.709 footage toward wider primaries, and a dedicated detail path restores fine texture in shadows and highlights that SDR compression flattens. You send an object-storage file, we return an HDR master.

Full API reference
Python
import requests

url = "https://api.mlslabs.io/v1/sdr2hdr/jobs"
resp = requests.post(
    url,
    headers={"X-API-Key": "YOUR_API_KEY"},
    json={
        "input": "s3://bucket/input.mp4",
        "output": "s3://bucket/output-hdr.mp4",
        "target": "hdr10",
        "preset": "cinema",
    },
)
job = resp.json()          # {"job_id": "...", "status": "queued"}
print(job["job_id"])

Key features

Why teams pick this API

Adaptive tone mapping

Per-scene luminance analysis lifts shadows and highlights into HDR range without halos or banding.

Wide-gamut recovery

Expands Rec.709 toward wider primaries with learned color relationships, keeping skin tones and brand colors intact.

Detail & texture restoration

A dedicated detail path rebuilds fine texture in shadows and highlights that SDR compression flattens.

Built for pipelines

Async job API, per-second billing, object-storage in/out, and webhook callbacks that scale with your backlog.

Use cases

Built for real workflows

Remaster legacy libraries

Upgrade archived SD/HD catalogs to HDR masters in one pass, without re-grading by hand.

Streaming & short-drama export

Deliver eye-catching HDR versions of new productions alongside SDR copies for every platform.

OTT master preparation

Produce consistent HDR10 masters ready for streaming and disc workflows.

Pricing

Credit-based pricing

Every request draws from your credit pool. This API bills 40 credits per second of input video.

Billing unitCost in creditsNotes
Processing time40 creditsPer second of input video
Failed jobs0 creditsCredits auto-refunded
Batch processingSame rateOne queue, per-file callbacks

See credit packs & full pricing

Learn

Guides, tutorials & comparisons

Browse all tutorials

FAQ

Common questions

Which HDR formats do you output?

HDR10 first, with HDR10+ and HLG on request. Metadata is generated per scene.

Do I need a matching HDR reference?

No. A single SDR source is enough — the model was trained on professionally graded SDR-to-HDR pairs.

Is it real-time?

It runs as an async job — typically a few minutes for a standard-length episode, scaling with your backlog.

Can I chain it with your other APIs?

Yes. SDR2HDR is designed to sit next to Super Resolution and Frame Interpolation in one enhancement chain.