Leveraging Compression to Construct Transferable Bitrate Ladders

2025-12-14

AI Summary

To address the poor generalizability of fixed bitrate ladders in adaptive streaming and the high computational overhead of video-specific convex hull construction, this paper proposes a source-video-aware, transferable bitrate ladder generation method. The core innovation lies in the first integration of compression process modeling with pre-compression feature extraction—specifically texture and motion characteristics—enabling cross-encoder-configuration transfer without per-video encoding or dedicated machine learning model training. Leveraging VMAF prediction and Bjøntegaard-delta optimization, our method is evaluated on a large-scale video dataset. It significantly outperforms both fixed ladders and state-of-the-art ML-based approaches: achieving an average 32% bitrate reduction while preserving QoE, and approaching the performance of exhaustive encoding-based optimal convex hulls.

Technology Category

• Machine Learning: Learning on the Edge & Model Compression
• Computer Vision: Learning & Optimization for CV
• Search and Optimization: Learning to Search

Application Category

• User Modeling, Personalization and Recommendation: Fairness-aware retrieval and ranking
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

Over the past few years, per-title and per-shot video encoding techniques have demonstrated significant gains as compared to conventional techniques such as constant CRF encoding and the fixed bitrate ladder. These techniques have demonstrated that constructing content-gnostic per-shot bitrate ladders can provide significant bitrate gains and improved Quality of Experience (QoE) for viewers under various network conditions. However, constructing a convex hull for every video incurs a significant computational overhead. Recently, machine learning-based bitrate ladder construction techniques have emerged as a substitute for convex hull construction. These methods operate by extracting features from source videos to train machine learning (ML) models to construct content-adaptive bitrate ladders. Here, we present a new ML-based bitrate ladder construction technique that accurately predicts the VMAF scores of compressed videos, by analyzing the compression procedure and by making perceptually relevant measurements on the source videos prior to compression. We evaluate the performance of our proposed framework against leading prior methods on a large corpus of videos. Since training ML models on every encoder setting is time-consuming, we also investigate how per-shot bitrate ladders perform under different encoding settings. We evaluate the performance of all models against the fixed bitrate ladder and the best possible convex hull constructed using exhaustive encoding with Bjontegaard-delta metrics.

Problem

• Constructing content-adaptive bitrate ladders for video encoding
• Reducing computational overhead of convex hull construction
• Predicting VMAF scores accurately through compression analysis

Innovation

• ML-based bitrate ladder construction technique
• Predicts VMAF scores from compression analysis
• Uses perceptual measurements on source videos