BOHA: A High Performance VSLAM Backend Optimization Hardware Accelerator Using Recursive Fine-Grain H-Matrix Decomposition and Early-Computing With Approximate Linear Solver.
Ye LiuKun HuangNeng ZhaoJingyuan LiShuang HaoZili HuangXiuyuan QiXinghao WangLiang ZhouLiang ChangJun ZhouPublished in: IEEE Trans. Circuits Syst. II Express Briefs (2023)
Keyphrases
- fine grain
- back end
- coarse grain
- distributed memory
- matrix decomposition
- parallel computation
- parallel implementation
- low rank matrix
- user friendly
- low rank
- nonnegative matrix factorization
- data management
- shared memory
- building blocks
- field programmable gate array
- massively parallel
- data sets
- nested transactions
- parallel processing
- low cost
- parallel machines
- database systems
- singular value decomposition
- parallel algorithm
- data types
- multithreading
- databases
- query processing
- data analysis