Action-Constrained Reinforcement Learning for Frame-Level Bit Allocation in HEVC/H.265 through Frank-Wolfe Policy Optimization.
Yung-Han HoYun LiangChia-Hao KaoWen-Hsiao PengPublished in: CoRR (2022)
Keyphrases
- bit allocation
- reinforcement learning
- video compression
- action selection
- action space
- optimal policy
- image compression
- video coding
- bit rate
- state action
- rate distortion
- low bit rate
- policy search
- macroblock
- state space
- rate control
- video quality
- motion compensation
- agent learns
- vector quantizer
- intra prediction
- reference frame
- video coding standard
- video codec
- computer vision
- video frames
- image processing