English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

In modern video encoders, rate control is a critical component and has beenheavily engineered. It decides how many bits to spend to encode each frame, inorder to optimize the rate-distortion trade-off over all video frames. This isa challenging constrained planning problem because of the complex dependencyamong decisions for different video frames and the bitrate constraint definedat the end of the episode.  We formulate the rate control problem as a Partially Observable MarkovDecision Process (POMDP), and apply imitation learning to learn a neural ratecontrol policy. We demonstrate that by learning from optimal video encodingtrajectories obtained through evolution strategies, our learned policy achievesbetter encoding efficiency and has minimal constraint violation. In addition toimitating the optimal actions, we find that additional auxiliary losses, dataaugmentation/refinement and inference-time policy improvements are critical forlearning a good rate control policy. We evaluate the learned policy against therate control policy in libvpx, a widely adopted open source VP9 codec library,in the two-pass variable bitrate (VBR) mode. We show that over a diverse set ofreal-world videos, our learned policy achieves 8.5% median bitrate reductionwithout sacrificing video quality.

Neural Rate Control for Video Encoding using Imitation Learning