Per clip adaptive Lagrangian multiplier optimisation with low resolution proxies

Ringis, Daniel; Pitié, François; Kokaram, Anil

doi:10.1117/12.2567654

Cited by 6 publications

(14 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In recently published works [13], [14] we introduce the use of direct optimisation to maximise BD-R and evaluated it on an expanded dataset, primarily based on the YouTube-UGC dataset [22]. The dataset was complemented with clips from Fig.…”

Section: B Direct Optimisation Of Kmentioning

confidence: 99%

“…Because of this, we have proposed in [14] to reduce the computational cost by using proxies (low resolutions, older codecs) to calculate k in practice. We were able to get a reduction of 3.1×, while achieving 83% of the gain found by direct optimisation.…”

Section: Machine-learning Prediction Of Kmentioning

confidence: 99%

“…We propose in this section to further explore the machine learning approach proposed in our previous work [14]. The main contributions are 1) to extend the feature set to per frame encoding statistics instead of global sequence statistics, 2) to include semantic visual features, and lastly 3) to employ a Deep Neural Network (DNN) instead of Random Forest.…”

Section: Deep Learning For Per Clip Lagrangian Multiplier Optimisationmentioning

confidence: 99%

“…This is because the statistics of video clips varies greatly over any corpus. In our previous work [13], [14], we have shown that direct optimisation methods can lead to bitrate gains per clip of up to 20%. Unfortunately, this bitrate savings came at a high computational cost, i.e.…”

Section: Introductionmentioning

confidence: 97%

See 3 more Smart Citations

Near Optimal Per-Clip Lagrangian Multiplier Prediction in HEVC

Ringis,

Pitié,

Kokaram

2022

Preprint

Self Cite

View full text Add to dashboard Cite

The majority of internet traffic is video content. This drives the demand for video compression to deliver high quality video at low target bitrates. Optimising the parameters of a video codec for a specific video clip (per-clip optimisation) has been shown to yield significant bitrate savings. In previous work we have shown that per-clip optimisation of the Lagrangian multiplier leads to up to 24% BD-Rate improvement. A key component of these algorithms is modeling the R-D characteristic across the appropriate bitrate range. This is computationally heavy as it usually involves repeated video encodes of the high resolution material at different parameter settings. This work focuses on reducing this computational load by deploying a NN operating on lower bandwidth features. Our system achieves BD-Rate improvement in approximately 90% of a large corpus with comparable results to previous work in direct optimisation.

show abstract

Section: B Direct Optimisation Of Kmentioning

confidence: 99%

Section: Machine-learning Prediction Of Kmentioning

confidence: 99%

Section: Deep Learning For Per Clip Lagrangian Multiplier Optimisationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 97%

See 2 more Smart Citations

Near Optimal Per-Clip Lagrangian Multiplier Prediction in HEVC

Ringis,

Pitié,

Kokaram

2022

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…motion vectors, block type (Skipped/Intra/Inter), and bit allocation at the frame and clip levels. 8 The general idea which we have applied in our recent work [9][10][11] is to adjust λ away from the codec default by using a constant k as follows,…”

Section: Introductionmentioning

confidence: 99%

Per-clip and per-bitrate adaptation of the Lagrangian multiplier in video coding

Ringis,

Pitié,

Kokaram

2022

Preprint

Self Cite

View full text Add to dashboard Cite

In the past ten years there have been significant developments in optimization of transcoding parameters on a per-clip rather than per-genre basis. In our recent work we have presented per-clip optimization for the Lagrangian multiplier in Rate controlled compression, which yielded BD-Rate improvements of approximately 2% across a corpus of videos using HEVC. However, in a video streaming application, the focus is on optimizing the rate/distortion tradeoff at a particular bitrate and not on average across a range of performance.We observed in previous work that a particular multiplier might give BD rate improvements over a certain range of bitrates, but not the entire range. Using different parameters across the range would improve gains overall. Therefore here we present a framework for choosing the best Lagrangian multiplier on a per-operating point basis across a range of bitrates. In effect, we are trying to find the para-optimal gain across bitrate and distortion for a single clip. In the experiments presented we employ direct optimization techniques to estimate this Lagrangian parameter path approximately 2,000 video clips. The clips are primarily from the YouTube-UGC dataset. We optimize both for bitrate savings as well as distortion metrics (PSNR, SSIM).

show abstract

Frame-Type Sensitive RDO Control for Content-Adaptive Encoding

Vibhoothi

Pitié

Kokaram

2022

2022 IEEE International Conference on Image Processing (ICIP)

View full text Add to dashboard Cite

Video transcoding is an increasingly important application in the streaming media industry. It has become important to investigate the optimisation of transcoder parameters for a single clip simply because of the immense number of playbacks for popular clips. In this paper, we explore the use of a canned optimiser to estimate the optimal Rate-Distortion (RD) tradeoff achievable for a particular clip. We show that by adjusting the Lagrange multiplier in RD optimisation on keyframes alone we can achieve more than 10× the previous BD-Rate gains possible without affecting quality for any operating point.

show abstract

Per clip adaptive Lagrangian multiplier optimisation with low resolution proxies

Cited by 6 publications

References 0 publications

Near Optimal Per-Clip Lagrangian Multiplier Prediction in HEVC

Near Optimal Per-Clip Lagrangian Multiplier Prediction in HEVC

Per-clip and per-bitrate adaptation of the Lagrangian multiplier in video coding

Frame-Type Sensitive RDO Control for Content-Adaptive Encoding

Contact Info

Product

Resources

About