Learned Scalable Video Coding For Humans and Machines

Description

An end-to-end learnable video codec that supports machine vision tasks in its base layer and input reconstruction for human viewing in its enhancement layer. It introduces a new paradigm for video coding that efficiently represents and compresses video for both machine and human use.

URL

• Visit resource

Related resources

• MIT 6.441 Information Theory — OCW Lecture NotesMIT OpenCourseWare graduate information theory course covering lossless/lossy coding foundations underlying video codec design.
• Stanford CS348K: Video Compression (Traditional and Learned)Graphics/systems course lecture covering both classic block-based codecs and learned neural video compression, public slides.
• MirrorFly IM SoftwareMirrorFly is a self-hosted instant messaging software that enables businesses to create a white-label team chat and meeting app. The soluti…
• Building Service Topology at Scale (Netflix Tech Blog)Real-time service dependency-mapping system for streaming microservices at scale, architecture and lessons learned.
• The Broadcast Knowledge — OTT Talks ArchiveArchive of recorded conference talks and webinars on OTT streaming and codec engineering topics.
• Building Your Own Scalable Video Streaming Server (Part 1)Code-along tutorial series building a microservices-based video streaming server with FFmpeg HLS segmenting.
• A Hitchhiker's Guide to Structural SimilarityNetflix-affiliated deep-dive paper on SSIM quality metrics and their application to streaming QoE evaluation.
• Video Encoding to AV1 Guide (WIP)Long-running community-maintained guide to AV1 software encoding (libaom/aomenc) on the Level1Techs forum.
• Building Your First Video Pipeline: FFmpeg & MediaMTX Basics (Part 1)Technical tutorial walking through demux→decode→filter→encode→mux pipeline stages toward building a low-latency streaming setup with FFmpeg…
• Fora Soft: Video Streaming Deep-Dive GuideFree 9-chapter technical guide covering HLS, DASH, WebRTC, CMAF, CDN, and DRM fundamentals.
• Ronald S. Bultje — Random thoughtsPersonal blog by ex-Google VP9 lead dev, dav1d/Eve encoder co-creator, Two Orioles founder. Deep VP9/AV1 codec internals writeups.
• MDN: Media Source Extensions APIOfficial MDN reference documentation for the Media Source Extensions API (MediaSource, SourceBuffer, SourceBufferList) used to build custom…
• Encoding Bitrate Optimization Using Playback Statistics for HTTP-based Adaptive Video StreamingAcademic paper with industry framing on ABR bitrate ladder optimization using real playback statistics.
• High Quality Video Encoding at Scale (Netflix Tech Blog)Deep-dive into Netflix's EC2-based elastic encode pipeline, per-title encoding, and automated quality scoring.
• VisualOn + Axinom Joint Whitepaper: Encoding Cost Optimization2025 joint technical whitepaper covering encoding cost optimization across DASH/HLS/CMAF and DRM (Widevine/PlayReady/FairPlay) tradeoffs.