V-Nova LCEVC tackles video messaging quality via on-device layered encoding
V-Nova details how MPEG-5 LCEVC enables scalable, two-layer video encoding directly on sender devices for video messaging, improving quality without requiring cloud transcoding. This approach allows services to deliver a universal baseline while offering higher quality to recipients with LCEVC-enabled apps. The technology can be deployed via app updates on iOS and Android, working with existing AVC/HEVC hardware encoders.
Key Takeaways
- Software-level deployment on iOS and Android enables LCEVC via app updates without requiring device firmware or OS-level codec changes
- On-device encoding uses existing hardware to produce a baseline stream (e.g., 360p at 1.5 Mbps) while adding a lightweight enhancement layer
- Testing on 50 user-generated content clips demonstrated 720p reconstruction at under 3 Mbps with visible quality improvements over the base stream
- Scalable delivery reduces Time-To-First-Frame (TTFF) by prioritizing the base stream and applying enhancements dynamically as bandwidth permits
Why It Matters
In edge-processed video workflows like private messaging, cloud-based re-encoding is often skipped to preserve privacy and reduce latency, locking users into low-resolution 'lowest common denominator' files. LCEVC bypasses this limitation by building scalability into the initial capture on the sender’s handset. This shift allows platforms to improve Quality of Experience (QoE) asymmetrically: capable recipients see sharp video, while legacy users maintain standard playback. As social and messaging apps face pressure to reduce egress costs and power consumption, this codec-agnostic 'enhancement' model provides a path to higher fidelity without the 5–7 year wait typically required for hardware-native codec cycles. Watch for LCEVC adoption rates in high-scale social platforms like Facebook or WhatsApp following these mobile SDK optimizations.
Additional Context
V-Nova has aggressively positioned MPEG-5 LCEVC as a bridge for next-generation standards. In April 2026, the company demonstrated the technology at NAB Show to highlight its role in Brazil’s DTV+ (TV 3.0) transition, where LCEVC is being used alongside VVC to deliver 4K HDR at under 10 Mbps. Per Broadcast Beat (April 2026), the Brazil rollout represents the first national mandate for the technology, providing a critical commercial proof point for the standard’s scalability across both broadcast and IP-based delivery. At the same event, V-Nova highlighted its growing ecosystem of more than 50 partners, including integrations with open-source projects like FFmpeg and Shaka Player, and hardware demonstrations with Realtek and Harmonic. On the regulatory and licensing front, V-Nova launched two distinct licensing programs in August 2025 aimed at accelerating global adoption. According to TV Technology (August 2025), one program targets video distributors with royalties based on user counts—featuring an $8 million annual cap for early adopters—while the second charges device manufacturers a per-unit fee starting at $0.20. These programs are specifically designed to support the technology's inclusion in both the American ATSC 3.0 standard and Brazil's TV 3.0. Additionally, V-Nova joined the Access Advance HEVC patent pool in late 2025, a strategic move intended to simplify the IP landscape for platforms already utilizing the HEVC base codec alongside LCEVC enhancements. Technically, LCEVC’s appeal lies in its efficiency compared to standalone next-gen codecs. Independent studies published in early 2026 by netint.com noted that LCEVC can provide bitrate savings of 30% to 55% across H.264 and HEVC while reducing encoding complexity by up to 3.6x compared to full-resolution alternatives. This low-complexity profile is increasingly valuable for edge-AI applications; as reported by edge-ai-vision.com in April 2026, the layered structure allows AI triage systems to process the lower-resolution base layer for inference while only pulling the enhancement layer for high-precision recognition tasks.
Read full article at v-nova.com