Automatically scheduling halide image processing pipelines
Ravi Teja Mullapudi, Andrew Adams, Dillon Sharlet et al. · ACM Transactions on Graphics · 2016
The Halide image processing language has proven to be an effective system for authoring high-performance image processing code. Halide programmers need only provide a high-level strategy for mapping an image processing pipeline to a parallel machine (a schedule ), and the Halide compiler carries out the mechanical task of generating platform-specific code that implements the schedule. Unfortunately, designing high-performance schedules for complex image processing pipelines requires substantial knowledge of modern hardware architecture and code-optimization techniques. In this paper we provide an algorithm for automatically generating high-performance schedules for Halide programs. Our solution extends the function bounds analysis already present in the Halide compiler to automatically perform locality and parallelism-enhancing global program transformations typical of those employed by expert Halide developers. The algorithm does not require costly (and often impractical) auto-tuning, and, in seconds, generates schedules for a broad set of image processing benchmarks that are performance-competitive with, and often better than, schedules manually authored by expert Halide developers on server and mobile CPUs, as well as GPUs.
Cited by 174 works. doi:10.1145/2897824.2925952
Authors
References
- ImageNet classification with deep convolutional neural networks
- Very Deep Convolutional Networks for Large-Scale Image Recognition
- Caffe
- A Combined Corner and Edge Detector
- Halide
- Fast cost-volume filtering for visual correspondence and beyond
- Real-time edge-aware image processing with the bilateral grid
- Real-time edge-aware image processing with the bilateral grid
- Local Laplacian filters
- Local Laplacian filters