TY - GEN
T1 - PixelPrune
T2 - 35th Edition of the Great Lakes Symposium on VLSI 2025, GLSVLSI 2025
AU - Mohammadi, Mohammadreza
AU - Morsali, Mehrdad
AU - Tabrizchi, Sepehr
AU - Reidy, Brendan
AU - Roohi, Arman
AU - Angizi, Shaahin
AU - Zand, Ramtin
N1 - Publisher Copyright:
© 2025 Copyright held by the owner/author(s).
PY - 2025/6/29
Y1 - 2025/6/29
N2 - This paper proposes PixelPrune, an approach to address two primary challenges in artificial intelligence of things (AIoT) vision systems: (1) the energy-intensive analog-to-digital converters (ADCs) required in the sensing unit for converting analog pixel arrays to digital tensors, and (2) the high data transfers between the sensing unit and computing unit. Our proposed solution involves the implementation of an in-sensor binary segmentation model on analog memristive crossbars to identify the important pixels and prune out the background information. Additionally, we propose a data transfer scheme that adaptively selects between dense and sparse data transfer formats based on the sparsity ratio measured from the segmentation mask obtained by the segmentation model. Our results demonstrate that the proposed object detection system achieves significant energy savings along with a considerable up to 95% reduction in data transfer, all while maintaining high accuracy.
AB - This paper proposes PixelPrune, an approach to address two primary challenges in artificial intelligence of things (AIoT) vision systems: (1) the energy-intensive analog-to-digital converters (ADCs) required in the sensing unit for converting analog pixel arrays to digital tensors, and (2) the high data transfers between the sensing unit and computing unit. Our proposed solution involves the implementation of an in-sensor binary segmentation model on analog memristive crossbars to identify the important pixels and prune out the background information. Additionally, we propose a data transfer scheme that adaptively selects between dense and sparse data transfer formats based on the sparsity ratio measured from the segmentation mask obtained by the segmentation model. Our results demonstrate that the proposed object detection system achieves significant energy savings along with a considerable up to 95% reduction in data transfer, all while maintaining high accuracy.
KW - and artificial intelligence of things (AIoT)
KW - edge computing
KW - In-sensor computing
KW - machine learning (ML) systems
UR - https://www.scopus.com/pages/publications/105017553388
UR - https://www.scopus.com/inward/citedby.url?scp=105017553388&partnerID=8YFLogxK
U2 - 10.1145/3716368.3735177
DO - 10.1145/3716368.3735177
M3 - Conference contribution
AN - SCOPUS:105017553388
T3 - Proceedings of the ACM Great Lakes Symposium on VLSI, GLSVLSI
SP - 312
EP - 319
BT - GLSVLSI 2025 - Proceedings of the Great Lakes Symposium on VLSI 2025
PB - Association for Computing Machinery
Y2 - 30 June 2025 through 2 July 2025
ER -