UCL Discovery
UCL home » Library Services » Electronic resources » UCL Discovery

Escaping the complexity-bitrate-quality barriers of video encoders via deep perceptual optimization

Chadha, A; Anam, R; Fadeev, I; Giotsas, V; Andreopoulos, Y; (2020) Escaping the complexity-bitrate-quality barriers of video encoders via deep perceptual optimization. In: Tescher, A and Ebrahimi, T, (eds.) Proceedings of SPIE - The International Society for Optical Engineering. SPIE: Online conference. Green open access

[img]
Preview
Text
115100C.pdf - Published version

Download (685kB) | Preview

Abstract

We extend the concept of learnable video precoding (rate-aware neural-network processing prior to encoding) to deep perceptual optimization (DPO). Our framework comprises a pixel-to-pixel convolutional neural network that is trained based on the virtualization of core encoding blocks (block transform, quantization, block-based prediction) and multiple loss functions representing rate, distortion and visual quality of the virtual encoder. We evaluate our proposal with AVC/H.264 and AV1 under per-clip rate-quality optimization. The results show that DPO offers, on average, 14.2% bitrate reduction over AVC/H.264 and 12.5% bitrate reduction over AV1. Our framework is shown to improve both distortion- and perception-oriented metrics in a consistent manner, exhibiting only 3% outliers, which correspond to content with peculiar characteristics. Thus, DPO is shown to offer complexity-bitrate-quality tradeoffs that go beyond what conventional video encoders can offer

Type: Proceedings paper
Title: Escaping the complexity-bitrate-quality barriers of video encoders via deep perceptual optimization
Event: SPIE Optical Engineering + Applications, 2020
ISBN-13: 9781510638266
Open access status: An open access version is available from UCL Discovery
DOI: 10.1117/12.2567549
Publisher version: https://doi.org/10.1117/12.2567549
Language: English
Additional information: This version is the version of record. For information on re-use, please refer to the publisher’s terms and conditions.
Keywords: visual quality, neural networks, video coding, AVC/H.264, AV1
UCL classification: UCL
UCL > Provost and Vice Provost Offices
UCL > Provost and Vice Provost Offices > UCL BEAMS
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science
UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Electronic and Electrical Eng
URI: https://discovery.ucl.ac.uk/id/eprint/10113692
Downloads since deposit
74Downloads
Download activity - last month
Download activity - last 12 months
Downloads by country - last 12 months

Archive Staff Only

View Item View Item