Skip to search formSkip to main contentSkip to account menu

Halide: a language and compiler for optimizing parallelism, locality, and recomputation in image processing pipelines

A systematic model of the tradeoff space fundamental to stencil pipelines is presented, a schedule representation which describes concrete points in this space for each stage in an image processing pipeline, and an optimizing compiler for the Halide image processing language that synthesizes high performance implementations from a Halide algorithm and a schedule are presented.
View on ACM (opens in a new tab)

OpenTuner: An extensible framework for program autotuning

The efficacy and generality of OpenTuner are demonstrated by building autotuners for 7 distinct projects and 16 total benchmarks, showing speedups over prior techniques of these projects of up to 2.8χ with little programmer effort.
View on ACM (opens in a new tab)

Decoupling algorithms from schedules for easy optimization of image processing pipelines

This work proposes a representation for feed-forward imaging pipelines that separates the algorithm from its schedule, enabling high-performance without sacrificing code clarity, and demonstrates the power of this representation by expressing a range of recent image processing applications in an embedded domain specific language called Halide and compiling them for ARM, x86, and GPUs.
View on ACM (opens in a new tab)

Learning to optimize halide with tree search and random programs

This work presents a new algorithm to automatically schedule Halide programs for high-performance image processing and deep learning that produces schedules which are on average almost twice as fast as the existing Halide autoscheduler without autotuning, or more than two as fast with, and is the first automatic scheduling algorithm to significantly outperform human experts on average.
View on ACM (opens in a new tab)

DiffTaichi: Differentiable Programming for Physical Simulation

We present DiffTaichi, a new differentiable programming language tailored for building high-performance differentiable physical simulators. Based on an imperative programming language, DiffTaichi…
View PDF on arXiv (opens in a new tab)

Automatically scheduling halide image processing pipelines

An algorithm for automatically generating high-performance schedules for Halide programs that does not require costly (and often impractical) auto-tuning, and 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.
View on ACM (opens in a new tab)

Neural Kernels Without Tangents

Using well established feature space tools such as direct sum, averaging, and moment lifting, an algebra for creating "compositional" kernels from bags of features is presented that corresponds to many of the building blocks of "neural tangent kernels (NTK).
View PDF on arXiv (opens in a new tab)

Darkroom: compiling high-level image processing code into hardware pipelines

The semantics of the Darkroom language allow it to compile programs directly into line-buffered pipelines, with all intermediate values in local line-buffer storage, eliminating unnecessary communication with off-chip DRAM.
View via Publisher (opens in a new tab)

Programming Heterogeneous Systems from an Image Processing DSL

The image processing language Halide is extended so users can specify which portions of their applications should become hardware accelerators, and a compiler is provided that uses this code to automatically create the accelerator along with the “glue” code needed for the user’s application to access this hardware.
View on ACM (opens in a new tab)

Taichi: a language for high-performance computation on spatially sparse data structures

Fig. 1. (Top) We propose the Taichi programming language, which exposes a high-level interface for developing and processing spatially sparse multi-level data structures, and an optimizing compiler…
View via Publisher (opens in a new tab)
...
By clicking accept or continuing to use the site, you agree to the terms outlined in our Privacy Policy (opens in a new tab), Terms of Service (opens in a new tab), and Dataset License (opens in a new tab)