What Is Ultra HDR and How Does It Intersect With AVIF?

This article provides an overview of Google’s Ultra HDR image format, explaining its core architecture, standard dynamic range (SDR) backward compatibility, and how it utilizes the modern AVIF container. Readers will learn how gain map technology bridges legacy and modern display capabilities, and why encoding Ultra HDR data within AVIF represents a significant leap forward in visual quality and file efficiency over traditional JPEG implementations.

What Is the Ultra HDR Specification?

Introduced by Google alongside Android 14, the Ultra HDR image format brings high dynamic range photography to standard imaging ecosystems without breaking compatibility with legacy hardware and software.

Traditional HDR formats often face display mismatch issues: rendering a native HDR image directly on an SDR screen often leads to blown-out highlights, crushed shadows, or incorrect tone-mapping. Ultra HDR solves this by using a dual-layer approach built around an HDR gain map.

An Ultra HDR file consists of three essential components:

  1. A Base Image: Typically an 8-bit SDR image that standard viewers and legacy applications can display without any modifications.
  2. A Gain Map: A secondary, lower-resolution grayscale or color image that stores the difference in luminance and chrominance between the SDR base and the full HDR target.
  3. Metadata: Standardized mathematical parameters (aligned with ISO 21496-1) that instruct an HDR-capable display engine on how to blend the base image and the gain map according to the device's specific display brightness limits.

When opened on an SDR device, the decoder renders only the base image. When opened on an HDR display, compatible software reads the gain map and metadata, recalculating pixels to brighten highlights and expand color gamuts up to the target HDR peak brightness.

The Role of the AVIF Container

AVIF (AV1 Image File Format) is an open, royalty-free image format derived from the video codec AV1 and standardized by the Alliance for Open Media (AOMedia). It relies on the ISO Base Media File Format (ISOBMFF) container structure, which allows it to store complex media arrangements, including multiple image items, transparency (alpha channels), image sequences, and auxiliary metadata.

AVIF offers higher compression efficiency compared to legacy codecs like JPEG, reducing file sizes by up to 50% or more at comparable perceptual quality. Furthermore, AVIF natively supports 10-bit and 12-bit color depths and modern color spaces such as BT.2020 and DCI-P3, significantly reducing compression artifacts, color blocking, and banding in smooth gradients.

How Ultra HDR Intersects with AVIF

While Ultra HDR initially launched with JPEG as its primary transport mechanism, the specification was engineered to be format-agnostic. The intersection of Ultra HDR and AVIF leverages the structural advantages of the ISOBMFF container and the compression power of the AV1 codec.

1. Native Item Relationships

In a legacy JPEG, embedding an Ultra HDR gain map requires storing secondary images inside custom application markers (such as MPF or XMP blocks), which can cause parsing friction across different decoders. In contrast, the AVIF container natively supports primary and auxiliary items defined directly within the ISOBMFF standard. The primary SDR image and the gain map image are stored as discrete, addressable items, linked cleanly via standard item references (auxl or tone-mapping properties) and ISO 21496-1 metadata boxes.

2. Superior Gain Map and Base Compression

Both the base image and the gain map benefit from AV1 compression. Because gain maps represent relative luminance offsets across smooth gradients, traditional 8-bit JPEG compression can introduce spatial blockiness or stair-stepping that becomes visible when amplified on high-brightness HDR panels. AVIF allows the gain map itself—as well as the base image—to be compressed using 10-bit AV1, virtually eliminating banding in specular highlights and glowing areas.

3. Smaller Bandwidth Footprint

A dual-layer HDR image inherently contains more data than a standard single-layer image. When using JPEG, adding a gain map increases file size substantially. By wrapping Ultra HDR within AVIF, the combined size of the base image, gain map, and metadata is frequently smaller than a standard single-layer SDR JPEG of the same visual resolution.

Summary

The Ultra HDR specification standardizes how gain maps and tone-mapping metadata are packaged to ensure consistent rendering across both SDR and HDR displays. While JPEG provided immediate compatibility for the format's launch, the AVIF container provides the ideal architectural home for Ultra HDR—delivering native multi-item support, 10-bit precision, and high-efficiency compression needed for modern HDR workflows.