BONES: Near-Optimal Neural-Enhanced Video Streaming

BONES is a Neural-Enhanced Streaming (NES) control algorithm that jointly manages network and computational resources to maximize the quality of experience (QoE) for users. It formulates NES as a Lyapunov optimization problem and solves it online with near-optimal performance. BO

Link

• Visit resource

Related resources

• ksm — Go FairPlay Key Security ModuleGo implementation of an Apple FairPlay Streaming Key Security Module (KSM).
• CTA-WAVE dpctf-deploy — device test suiteCTA-WAVE Streaming Media Test Suite (Devices): MSE/EME/CMAF test content and local content server for device conformance.
• ffmpeg-gui (Haoming02)Lightweight Tauri FFmpeg GUI for batch-processing folders with CRF/bitrate control, progress bar, and interrupt support.
• rust-ac-ffmpegSafe minimal Rust bindings to libav with no bindgen dependency, supporting FFmpeg 4–7.
• Video Seal (Meta)Meta's efficient neural video and image forensic watermarking framework with a 256-bit payload model (MIT).
• Using DRM with Key Rotation (Unified Streaming docs)Official technical reference on CPIX ContentKeyPeriod/ContentKeyPeriodList spec detail and VOD vs live static/dynamic CPIX origin distincti…
• Codec Wiki (x266.mov)Community-maintained wiki covering AV1/HEVC/VVC codec internals, encoder settings, and tuning guides.