PipeFB: An Optimized Pipeline Parallelism Scheme to Reduce the Peak Memory Usage.
Wei JiangBo WangSheng MaXiang HouLibo HuangYi DaiJianbin FangPublished in: ICA3PP (2022)
Keyphrases
- memory usage
- memory footprint
- website
- memory requirements
- data sets
- detection scheme
- parallel processing
- learning scheme
- high image quality
- processing pipeline
- parallel execution
- parallel computation
- massively parallel
- significantly reduced
- image quality
- optimal solution
- multiscale
- image segmentation
- case study
- neural network