Uncertainty-Aware Deep Video Compression with Ensembles
This work addresses video compression efficiency for applications like streaming and storage, offering a novel approach to reduce bit rates, though it is incremental as it builds on existing two-stage models.
The paper tackled the problem of errors and artifacts in deep learning-based video compression caused by uncertainties in motion estimation and quantization, proposing an uncertainty-aware model with deep ensembles that achieved over 20% bit savings compared to DVC Pro on 1080p sequences.
Deep learning-based video compression is a challenging task, and many previous state-of-the-art learning-based video codecs use optical flows to exploit the temporal correlation between successive frames and then compress the residual error. Although these two-stage models are end-to-end optimized, the epistemic uncertainty in the motion estimation and the aleatoric uncertainty from the quantization operation lead to errors in the intermediate representations and introduce artifacts in the reconstructed frames. This inherent flaw limits the potential for higher bit rate savings. To address this issue, we propose an uncertainty-aware video compression model that can effectively capture the predictive uncertainty with deep ensembles. Additionally, we introduce an ensemble-aware loss to encourage the diversity among ensemble members and investigate the benefits of incorporating adversarial training in the video compression task. Experimental results on 1080p sequences show that our model can effectively save bits by more than 20% compared to DVC Pro.