NeMO: An Efficient Memory Access Architecture for Instant 3D Rendering
IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems · 2026-11 · 45 (11) · 5825-5838
DOI: 10.1109/tcad.2026.3674446 ↗Kurz gefasst
Fachartikel in der Zeitschrift IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 2026-11. Verfasst von Shangrong Li und 8 weiteren Autor:innen. Die Arbeit nennt 39 Literaturangaben. Gefördert von: National Natural Science Foundation of China, State Key Laboratory of Spatial Datum, Laboratory of Space Trusted Computing and Electronic Information Technology.
Abstract (Originalsprache)
Neural radiance fields (NeRFs) have emerged as a promising solution for 3-D spatial modeling, scene reconstruction, and AR/VR applications. However, NeRF rendering involves frequent and irregular off-chip memory accesses, particularly during ray sampling and hash encoding stages. Although existing edge platforms offer sufficient computational capability and memory bandwidth, such irregular memory access patterns lead to inefficient hardware utilization and significant latency overheads, making real-time rendering challenging. To address these issues, we present NeMO, an end-to-end accelerator specifically designed to enhance memory access efficiency in instant 3D rendering. NeMO incorporates three key innovations: 1) a novel occupancy grid compaction technique that completely eliminates off-chip memory accesses during the ray marching phase; 2) a hierarchical subgrid hashing strategy with block-by-block ray processing to optimize hash table access patterns, combined with a two-level cache architecture that reduces off-chip memory access by over 99% during multiresolution hash encoding; and 3) an end-to-end processor tailored for the NeRF pipeline, featuring fine-grained pipelining for ray marching and backpressure-enabled pipelining for multiresolution hash encoding. Implemented in a commercial 28-nm process technology, NeMO achieves real-time rendering at 79.23 FPS on synthetic datasets. It outperforms the Nvidia Jetson AGX Orin and high-end RTX 4080 graphics processing unit (GPU) by $51.78\times $ and $6.43\times $ in speed and $1767.08\times $ and $493.17\times $ in energy efficiency (EE), respectively. In addition, NeMO surpasses the latest NeRF accelerator, delivering $2.64\times $ higher speed and a $28.7\times $ off-chip bandwidth reduction, with per-frame memory access under 30 MB.
Förderung
- National Natural Science Foundation of China
- State Key Laboratory of Spatial Datum
- Laboratory of Space Trusted Computing and Electronic Information Technology
Zitieren
APA
Li, S., Liu, K., Shi, J., Guo, Z., Ding, C., Lin, Y., Song, Q., Wang, M., & Chen, Z. (2026). NeMO: An Efficient Memory Access Architecture for Instant 3D Rendering. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems, 45(11), 5825-5838. https://doi.org/10.1109/tcad.2026.3674446
Vancouver
Li S, Liu K, Shi J, Guo Z, Ding C, Lin Y, et al. NeMO: An Efficient Memory Access Architecture for Instant 3D Rendering. IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems. 2026;45(11):5825-5838. doi:10.1109/tcad.2026.3674446
BibTeX
@article{Li2026,
author = {Li, Shangrong and Liu, Kai and Shi, Jiangyi and Guo, Zibo and Ding, Chongyang and Lin, Yuna and Song, Qiuyu and Wang, Mengyuan and Chen, Zihang},
title = {NeMO: An Efficient Memory Access Architecture for Instant 3D Rendering},
journal = {IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems},
year = {2026},
volume = {45},
number = {11},
pages = {5825--5838},
doi = {10.1109/tcad.2026.3674446}
}
EUVIMED