Can AVIF Store Stereoscopic and 3D Spatial Images?

AVIF can store stereoscopic and 3D spatial photography within its container by utilizing the multi-image and auxiliary data mechanisms native to its underlying file architecture. While standard AVIF files typically contain single 2D images, the standard inherits specifications capable of packaging left-and-right stereoscopic pairs, depth maps, and spatial metadata. However, full implementation and broad playback support across web browsers and viewing software remain in the early stages of adoption.

Underlying Container Architecture

AVIF (AV1 Image File Format) is built upon the ISO Base Media File Format (ISOBMFF) and adheres to the Multi-Image Application Format (MIAF) framework. Because ISOBMFF was designed for complex multimedia, including video and multi-track audio, AVIF inherits a flexible container structure rather than a rigid, flat-file design.

This structure allows an AVIF container to store multiple discrete image items, reference relationships between items, and assign auxiliary roles to secondary data tracks.

Methods for Storing 3D and Spatial Data

There are two primary technical methods for packaging spatial photography into an AVIF container:

Multi-Item Stereoscopic Pairs: The container can encapsulate two distinct AV1-compressed image items—one for the left eye and one for the right eye. Using ISOBMFF entity groupings (groups) and item reference boxes (iref), the container specifies that both images are part of a synchronized stereoscopic pair intended for simultaneous rendering.
Color + Auxiliary Depth Maps: Instead of a full second angle, the file can contain a primary 2D image alongside an auxiliary image item representing depth or disparity information. This auxiliary item (commonly referenced via the auxl reference type) stores a grayscale depth map mapped pixel-by-pixel to the base image. Viewing devices can use this depth information to generate real-time parallax and simulated 3D spatial depth.

Standards and Spatial Metadata

For a device or application to correctly render a spatial photograph—especially in VR headsets or 3D displays—the container must declare camera geometry and stereo parameters. AVIF can store these parameters through standard metadata blocks, such as EXIF data or custom XML structures.

Standards organizations, including the Moving Picture Experts Group (MPEG) and the Alliance for Open Media (AOMedia), define conventions for identifying stereoscopic channels (e.g., left, right, or depth). AVIF uses these conventions directly within its metadata properties boxes to instruct the decoder on how to present the data.

Ecosystem and Decoder Limitations

Although the AVIF specification and its ISOBMFF foundation technically accommodate stereoscopic and spatial photography, real-world implementation faces limitations:

Encoder and Decoder Tooling: Most common encoding libraries (such as libavif) and standard consumer tools are optimized solely for single-frame 2D images or basic sequences (AVIS). Dedicated tooling to assemble multi-view or color-plus-depth AVIF files is specialized and less accessible.
Browser and Platform Support: Web browsers and standard operating systems typically parse only the primary item inside an AVIF container, ignoring auxiliary depth channels or secondary stereo views.
Comparison to HEIC: Spatial photography formats currently in mass consumer use—such as Apple's Spatial Photos—use HEIC (which also shares the ISOBMFF structure). While AVIF is technically capable of the exact same spatial arrangements using open-source AV1 compression, the commercial ecosystem has moved faster to standardize HEIC-based pipelines for 3D consumer hardware.