# Energy consumption reduction in the video chain through film grain removal/synthesis and generation of energy-aware images
_by Zoubida Ameur (IETR VAADER & InterDigital)_
###### tags: `VAADER` `Seminar`


## Abstract
The consumption of a video requires a considerable amount of energy during the various stages of its life-cycle. With a billion hours of video consumed daily, this contributes significantly to the greenhouse gas (GHG) emission. Therefore, reducing the end-to-end carbon footprint of the video chain, while preserving the quality of experience at the user side, is of high importance. In this presentation, we first address the reduction of energy consumption at the distribution level which translates into reducing bitrates. Given that the random nature of film grain makes it both difficult to preserve and very expensive to compress, removing it before encoding and synthesizing it after decoding helps to reduce bitrates while maintaining good visual quality. We present and compare the performance of our two proposed methods for film grain removal and synthesis: Deep-FG and Style-FG. Experiments show that fidelity to the reference grain, diversity of grain styles as well as a perceptually pleasant grain synthesis are achieved.

We then tackle the energy consumption reduction at all levels of the video chain using 3R-INN, an INN-based model that in addition to removing film grain, rescales the content and reduces its power consumption when displayed by some reduction rate R. While saving energy along the video chain, 3R-INN also provides a visually-pleasant content intended to be displayed with the possibility to restore either the HR grainy original content or a grain-free version. Experiments show that, while enabling significant energy savings for encoding (78%), decoding (77%) and rendering (5% to 20%), 3R-INN outperforms state-of-the-art film grain synthesis and energy-aware methods and achieves state-of-the art performance on the rescaling task on different test-sets.