Matches in SemOpenAlex for { <https://semopenalex.org/work/W4254778071> ?p ?o ?g. }
- W4254778071 endingPage "443" @default.
- W4254778071 startingPage "429" @default.
- W4254778071 abstract "This paper presents the design and implementation of PolyMage, a domain-specific language and compiler for image processing pipelines. An image processing pipeline can be viewed as a graph of interconnected stages which process images successively. Each stage typically performs one of point-wise, stencil, reduction or data-dependent operations on image pixels. Individual stages in a pipeline typically exhibit abundant data parallelism that can be exploited with relative ease. However, the stages also require high memory bandwidth preventing effective utilization of parallelism available on modern architectures. For applications that demand high performance, the traditional options are to use optimized libraries like OpenCV or to optimize manually. While using libraries precludes optimization across library routines, manual optimization accounting for both parallelism and locality is very tedious. The focus of our system, PolyMage, is on automatically generating high-performance implementations of image processing pipelines expressed in a high-level declarative language. Our optimization approach primarily relies on the transformation and code generation capabilities of the polyhedral compiler framework. To the best of our knowledge, this is the first model-driven compiler for image processing pipelines that performs complex fusion, tiling, and storage optimization automatically. Experimental results on a modern multicore system show that the performance achieved by our automatic approach is up to 1.81x better than that achieved through manual tuning in Halide, a state-of-the-art language and compiler for image processing pipelines. For a camera raw image processing pipeline, our performance is comparable to that of a hand-tuned implementation." @default.
- W4254778071 created "2022-05-12" @default.
- W4254778071 creator A5002285363 @default.
- W4254778071 creator A5006412542 @default.
- W4254778071 creator A5025342111 @default.
- W4254778071 date "2015-03-14" @default.
- W4254778071 modified "2023-10-01" @default.
- W4254778071 title "PolyMage" @default.
- W4254778071 cites W1779740778 @default.
- W4254778071 cites W1989988531 @default.
- W4254778071 cites W1993465480 @default.
- W4254778071 cites W1995270665 @default.
- W4254778071 cites W2003798513 @default.
- W4254778071 cites W2013680239 @default.
- W4254778071 cites W2032309817 @default.
- W4254778071 cites W2034761517 @default.
- W4254778071 cites W2035430533 @default.
- W4254778071 cites W2036959650 @default.
- W4254778071 cites W2051305716 @default.
- W4254778071 cites W2055312318 @default.
- W4254778071 cites W2063249715 @default.
- W4254778071 cites W2063656563 @default.
- W4254778071 cites W2070961300 @default.
- W4254778071 cites W2071110673 @default.
- W4254778071 cites W2082352135 @default.
- W4254778071 cites W2100218206 @default.
- W4254778071 cites W2104329103 @default.
- W4254778071 cites W2111308925 @default.
- W4254778071 cites W2134408405 @default.
- W4254778071 cites W2143425433 @default.
- W4254778071 cites W2143609451 @default.
- W4254778071 cites W2151764765 @default.
- W4254778071 cites W2153709524 @default.
- W4254778071 cites W2154078685 @default.
- W4254778071 cites W2154968583 @default.
- W4254778071 cites W2158626113 @default.
- W4254778071 cites W2161790807 @default.
- W4254778071 cites W2165100134 @default.
- W4254778071 cites W2172073571 @default.
- W4254778071 cites W3139167831 @default.
- W4254778071 cites W4235662649 @default.
- W4254778071 doi "https://doi.org/10.1145/2775054.2694364" @default.
- W4254778071 hasPublicationYear "2015" @default.
- W4254778071 type Work @default.
- W4254778071 citedByCount "17" @default.
- W4254778071 countsByYear W42547780712015 @default.
- W4254778071 countsByYear W42547780712017 @default.
- W4254778071 countsByYear W42547780712018 @default.
- W4254778071 countsByYear W42547780712019 @default.
- W4254778071 countsByYear W42547780712020 @default.
- W4254778071 countsByYear W42547780712021 @default.
- W4254778071 countsByYear W42547780712022 @default.
- W4254778071 countsByYear W42547780712023 @default.
- W4254778071 crossrefType "journal-article" @default.
- W4254778071 hasAuthorship W4254778071A5002285363 @default.
- W4254778071 hasAuthorship W4254778071A5006412542 @default.
- W4254778071 hasAuthorship W4254778071A5025342111 @default.
- W4254778071 hasConcept C115537543 @default.
- W4254778071 hasConcept C115961682 @default.
- W4254778071 hasConcept C118524514 @default.
- W4254778071 hasConcept C127413603 @default.
- W4254778071 hasConcept C154945302 @default.
- W4254778071 hasConcept C169590947 @default.
- W4254778071 hasConcept C173608175 @default.
- W4254778071 hasConcept C175309249 @default.
- W4254778071 hasConcept C199360897 @default.
- W4254778071 hasConcept C2781172179 @default.
- W4254778071 hasConcept C41008148 @default.
- W4254778071 hasConcept C43521106 @default.
- W4254778071 hasConcept C61483411 @default.
- W4254778071 hasConcept C87717796 @default.
- W4254778071 hasConcept C9417928 @default.
- W4254778071 hasConceptScore W4254778071C115537543 @default.
- W4254778071 hasConceptScore W4254778071C115961682 @default.
- W4254778071 hasConceptScore W4254778071C118524514 @default.
- W4254778071 hasConceptScore W4254778071C127413603 @default.
- W4254778071 hasConceptScore W4254778071C154945302 @default.
- W4254778071 hasConceptScore W4254778071C169590947 @default.
- W4254778071 hasConceptScore W4254778071C173608175 @default.
- W4254778071 hasConceptScore W4254778071C175309249 @default.
- W4254778071 hasConceptScore W4254778071C199360897 @default.
- W4254778071 hasConceptScore W4254778071C2781172179 @default.
- W4254778071 hasConceptScore W4254778071C41008148 @default.
- W4254778071 hasConceptScore W4254778071C43521106 @default.
- W4254778071 hasConceptScore W4254778071C61483411 @default.
- W4254778071 hasConceptScore W4254778071C87717796 @default.
- W4254778071 hasConceptScore W4254778071C9417928 @default.
- W4254778071 hasIssue "4" @default.
- W4254778071 hasLocation W42547780711 @default.
- W4254778071 hasOpenAccess W4254778071 @default.
- W4254778071 hasPrimaryLocation W42547780711 @default.
- W4254778071 hasRelatedWork W1541585229 @default.
- W4254778071 hasRelatedWork W1583465708 @default.
- W4254778071 hasRelatedWork W1601646354 @default.
- W4254778071 hasRelatedWork W2023505575 @default.
- W4254778071 hasRelatedWork W2039534605 @default.
- W4254778071 hasRelatedWork W2348711589 @default.
- W4254778071 hasRelatedWork W2354407271 @default.