Virtues of Recycling in Multi-rate Encoding
summary
Claim: double the x265 encoding speed while keeping the same encoder. Method: a hierarchical pipeline encodes low resolution or higher‑bitrate copies first, exports and scales motion vectors and quadtree structure into 2n×2n higher resolution encodes, then performs limited refinement. Built into x265 stable branch (v2.7+), implemented in the open source HEVC encoder, and validated with up to 250% speed-ups on higher resolutions with no visual quality penalty, and applicable to any motion‑compensated codec.
AI-written summary · about
chapters (7)
- 0:00Introduction
- 0:56Motivation
- 2:05Solution: Recycling Encoder Decisions
- 3:16Refining scaled information WHY REFINE
- 5:38Twice the speed, no quality loss
- 6:47Encode graphs EXAMPLE
- 8:56Summary
related talks
12mWhy is your encoder so slow? The curse of next-gen standards
30mA paradigm shift in codec standards: MPEG-5 Part 2 LCEVC, higher quality with any...
22mAI Enhanced GPU Video Coding: Achieving Joint High Compression Efficiency and Throughput
20mBoosting Video Codec With RISC-V Vector Extension
27mHighly-Efficient SVT-AV1-based Solutions for VOD Applications
19mRethinking the Video Transcoding Pipeline: How Pre-Filtering Holds the Key to Efficiency and Quality
11mFrom Artifacts to Art: A Data-Driven Guide to Android Video Encoding
10mHEVC Upload Experiments