Matches in SemOpenAlex for { <https://semopenalex.org/work/W4387250494> ?p ?o ?g. }
Showing items 1 to 80 of
80
with 100 items per page.
- W4387250494 endingPage "11" @default.
- W4387250494 startingPage "1" @default.
- W4387250494 abstract "Low-Power Edge-AI capabilities are essential for on-device extended reality (XR) applications to support the vision of Metaverse. In this work, we investigate two representative XR workloads: (i) Hand detection and (ii) Eye segmentation, for hardware design space exploration. For both applications, we train deep neural networks and analyze the impact of quantization and hardware-specific bottlenecks. Through simulations, we evaluate a CPU and two systolic inference accelerator implementations. Next, we compare these hardware solutions with advanced technology nodes. The impact of integrating state-of-the-art emerging non-volatile memory (NVM) technology (STT/SOT/VGSOT MRAM) into the XR-AI inference pipeline is evaluated. We found that significant energy benefits (≥24%) can be achieved for hand detection (IPS=10) and eye segmentation (IPS=0.1) by introducing NVM in the memory hierarchy for designs at 7nm node while meeting minimum IPS (inference per second). Moreover, we can realize substantial reduction in area (≥30%) owing to the small form factor of MRAM." @default.
- W4387250494 created "2023-10-03" @default.
- W4387250494 creator A5008011360 @default.
- W4387250494 creator A5053194542 @default.
- W4387250494 creator A5066176656 @default.
- W4387250494 creator A5072122272 @default.
- W4387250494 creator A5072539515 @default.
- W4387250494 creator A5088965878 @default.
- W4387250494 date "2023-01-01" @default.
- W4387250494 modified "2023-10-03" @default.
- W4387250494 title "Exploring Memory-Oriented Design Optimization of Edge-AI Hardware for Extended Reality Applications" @default.
- W4387250494 doi "https://doi.org/10.1109/mm.2023.3321249" @default.
- W4387250494 hasPublicationYear "2023" @default.
- W4387250494 type Work @default.
- W4387250494 citedByCount "0" @default.
- W4387250494 crossrefType "journal-article" @default.
- W4387250494 hasAuthorship W4387250494A5008011360 @default.
- W4387250494 hasAuthorship W4387250494A5053194542 @default.
- W4387250494 hasAuthorship W4387250494A5066176656 @default.
- W4387250494 hasAuthorship W4387250494A5072122272 @default.
- W4387250494 hasAuthorship W4387250494A5072539515 @default.
- W4387250494 hasAuthorship W4387250494A5088965878 @default.
- W4387250494 hasConcept C111919701 @default.
- W4387250494 hasConcept C113775141 @default.
- W4387250494 hasConcept C115537543 @default.
- W4387250494 hasConcept C118524514 @default.
- W4387250494 hasConcept C13164978 @default.
- W4387250494 hasConcept C138236772 @default.
- W4387250494 hasConcept C149635348 @default.
- W4387250494 hasConcept C154945302 @default.
- W4387250494 hasConcept C173608175 @default.
- W4387250494 hasConcept C177950962 @default.
- W4387250494 hasConcept C199360897 @default.
- W4387250494 hasConcept C2776214188 @default.
- W4387250494 hasConcept C2776221188 @default.
- W4387250494 hasConcept C2778100165 @default.
- W4387250494 hasConcept C41008148 @default.
- W4387250494 hasConcept C42935608 @default.
- W4387250494 hasConcept C43521106 @default.
- W4387250494 hasConcept C79974875 @default.
- W4387250494 hasConcept C89600930 @default.
- W4387250494 hasConcept C9390403 @default.
- W4387250494 hasConceptScore W4387250494C111919701 @default.
- W4387250494 hasConceptScore W4387250494C113775141 @default.
- W4387250494 hasConceptScore W4387250494C115537543 @default.
- W4387250494 hasConceptScore W4387250494C118524514 @default.
- W4387250494 hasConceptScore W4387250494C13164978 @default.
- W4387250494 hasConceptScore W4387250494C138236772 @default.
- W4387250494 hasConceptScore W4387250494C149635348 @default.
- W4387250494 hasConceptScore W4387250494C154945302 @default.
- W4387250494 hasConceptScore W4387250494C173608175 @default.
- W4387250494 hasConceptScore W4387250494C177950962 @default.
- W4387250494 hasConceptScore W4387250494C199360897 @default.
- W4387250494 hasConceptScore W4387250494C2776214188 @default.
- W4387250494 hasConceptScore W4387250494C2776221188 @default.
- W4387250494 hasConceptScore W4387250494C2778100165 @default.
- W4387250494 hasConceptScore W4387250494C41008148 @default.
- W4387250494 hasConceptScore W4387250494C42935608 @default.
- W4387250494 hasConceptScore W4387250494C43521106 @default.
- W4387250494 hasConceptScore W4387250494C79974875 @default.
- W4387250494 hasConceptScore W4387250494C89600930 @default.
- W4387250494 hasConceptScore W4387250494C9390403 @default.
- W4387250494 hasLocation W43872504941 @default.
- W4387250494 hasOpenAccess W4387250494 @default.
- W4387250494 hasPrimaryLocation W43872504941 @default.
- W4387250494 hasRelatedWork W1732210391 @default.
- W4387250494 hasRelatedWork W2110674252 @default.
- W4387250494 hasRelatedWork W2940862705 @default.
- W4387250494 hasRelatedWork W3143317418 @default.
- W4387250494 hasRelatedWork W3147787617 @default.
- W4387250494 hasRelatedWork W4235535008 @default.
- W4387250494 hasRelatedWork W4378806073 @default.
- W4387250494 hasRelatedWork W4383176055 @default.
- W4387250494 hasRelatedWork W4383501779 @default.
- W4387250494 hasRelatedWork W4386762860 @default.
- W4387250494 isParatext "false" @default.
- W4387250494 isRetracted "false" @default.
- W4387250494 workType "article" @default.