Hardware Race for Scalable Real-Time 3D Neural Reconstruction and Rendering
Year: 2025
Type: article
Abstract: Advances in 3D neural reconstruction and rendering promise to make high-fidelity 3D intelligence practical across various domains. Achieving this vision requires tackling two fundamental architectural challenges, memory irregularity and pipeline heterogeneity, that strain traditional accelerators.
Source: Computer
Authors: Sixu Li, Chaojian Li, Yingyan Celine Lin
Institution: Georgia Institute of Technology
Language: English
Subfield: Hardware and Architecture
Field: Computer Science
Domain: Physical Sciences
Open Access status: hybrid
Estimated APC paid: $2,995
OpenAlex ID: W7083315293