VideoTime3: A 40-Uj/frame 38 FPS Video Understanding Accelerator with Real-Time DiffFrame Temporal Redundancy Reduction and Temporal Modeling

Miaorong Wang,Yujun Lin,Zhekai Zhang,Ji Lin,Song Han,Anantha P. P. Chandrakasan

IEEE SOLID-STATE CIRCUITS LETTERS（2023）

引用 0|浏览105

关键词

Random access memory,Convolution,Generators,Real-time systems,Streaming media,Solid state circuits,Redundancy,Activation sparsity handling,deep learning (DL),Index Terms,hardware accelerator,video understanding

AI 理解论文

溯源树

样例

生成溯源树，研究论文发展脉络

Chat Paper

正在生成论文摘要