Description
Video-based Dynamic Mesh Coding (V-DMC) is a 3GPP media codec specification for compressing dynamic 3D mesh content, i.e., mesh geometry and attributes that change over time. It was introduced in Release 19 as part of the broader immersive media work in 3GPP SA4 and is documented in TR 26.813 and TS 26.956. The core idea is to convert the time-varying mesh into a set of 2D video sequences (geometry maps, attribute maps, and connectivity information) that can be encoded with conventional video codecs such as HEVC or VVC, and then reconstruct the mesh at the decoder. This leverages the maturity and hardware support of existing video codecs while providing efficient compression for dynamic 3D content.
The encoding pipeline begins with mesh preprocessing: the input mesh sequence is parameterized and mapped onto a 2D atlas, producing geometry and attribute videos. Connectivity information (how vertices are connected into triangles) is encoded separately, often using a base mesh plus displacement or refinement data. The resulting video streams are encoded with a standard video encoder, and the bitstreams are multiplexed into a V-DMC bitstream along with metadata describing the atlas layout, frame timing, and reconstruction parameters. At the decoder, the video streams are decoded, the atlas is unpacked, and the mesh is reconstructed frame by frame.
V-DMC supports both lossy and lossless coding modes and can operate at various levels of detail, making it suitable for adaptive streaming. It includes tools for temporal prediction of mesh motion, which exploits the high correlation between consecutive mesh frames to reduce bitrate. The codec also defines a set of profiles and levels to allow interoperability across devices with different decoding capabilities, from mobile AR glasses to high-end VR headsets. Because the underlying video codecs are widely deployed, V-DMC can be implemented with relatively modest additional hardware compared to dedicated mesh codecs.
In the 3GPP ecosystem, V-DMC is part of the media services framework that includes 5G Media Streaming (5GMS), immersive teleconferencing, and extended reality (XR) services. It is designed to work with DASH-based delivery, where mesh sequences are segmented and streamed adaptively according to network conditions. The codec's video-based nature also simplifies integration with existing content delivery networks and media players, since the transport and packaging can reuse established video streaming infrastructure.
Key components of V-DMC include the mesh preprocessor, the atlas generator, the video encoder(s), the bitstream multiplexer, and the decoder with mesh reconstruction. The specification defines conformance points and reference software to ensure consistent behavior across implementations. V-DMC is expected to be a foundational codec for 5G immersive services, enabling realistic avatars, volumetric video, and interactive 3D experiences over mobile networks.
Purpose & Motivation
V-DMC was created to address the lack of an efficient, standardized codec for dynamic 3D mesh content in 3GPP media services. Earlier immersive media work focused on static meshes or point clouds, but applications such as volumetric video, AR/VR avatars, and free-viewpoint video require time-varying meshes with high frame rates and detailed geometry. Existing mesh codecs were either proprietary, inefficient, or not designed for streaming over mobile networks.
The video-based approach was motivated by the ubiquity of hardware video decoders. By mapping mesh data to 2D videos, V-DMC can reuse HEVC/VVC decoding pipelines, reducing implementation cost and power consumption on mobile devices. This also allows the codec to benefit from ongoing advances in video compression, such as those standardized in VVC, without requiring a separate mesh-specific coding toolchain.
V-DMC also solves the problem of adaptive streaming for 3D content. Its support for temporal prediction, levels of detail, and DASH-compatible packaging enables service providers to deliver immersive experiences over variable network conditions. Introduced in Release 19, V-DMC is positioned to support the next generation of 5G XR and metaverse services, where dynamic mesh content is expected to be a primary media type.
Release Timeline
Evolution Across Releases
An overview written from the defining specifications, not a verified change history. For the exact text, open the specifications listed below.
Defining Specifications
3GPP specifications that define or reference V-DMC, with the latest known release. Sourced from the 3GPP document catalog — see methodology.
| Specification | Title | Release |
|---|---|---|
| TS 26.813 vk00 | Avatar Representation and Communication | Rel-20 read |
| TR 26.956 vj01 | Beyond 2D Video Formats & Codecs Study | Rel-19 read |