Row-Segmented Sparse-Dense Matrix Matrix Multiplication on GPUs.
Guoqing XiaoZiming LongYuedan ChenYunchuan QinFan WuPublished in: SmartWorld/UIC/ScalCom/DigitalTwin/PriComp/Meta (2022)
Keyphrases
- matrix multiplication
- message passing
- binary matrices
- motion field estimation
- dense stereo
- dense optical flow
- matrix factorization
- distributed memory
- optical flow
- sparse representation
- binary matrix
- high dimensional
- sparse matrix
- general purpose
- rows and columns
- dense motion estimation
- data matrix
- parallel programming
- image sequences
- tensor factorization
- coefficient matrix
- shared memory
- distributed systems
- pairwise
- lower bound
- three dimensional
- image segmentation
- computer vision