Abstract
Instant on-device Neural Radiance Fields (NeRFs) are in growing demand for unleashing the promise of immersive AR/VR experiences, but are still limited by their prohibitive training time. Our profiling analysis reveals a memory-bound inefficiency in NeRF training. To tackle this inefficiency, near-memory processing (NMP) promises to be an effective solution, but also faces challenges due to the unique workloads of NeRFs, including the random hash table lookup, random point processing sequence, and heterogeneous bottleneck steps. Therefore, we propose the first NMP framework, Instant-NeRF, dedicated to enabling instant on-device NeRF training. Experiments on eight datasets consistently validate the effectiveness of Instant-NeRF.
Original language | English (US) |
---|---|
Title of host publication | 2023 60th ACM/IEEE Design Automation Conference, DAC 2023 |
Publisher | Institute of Electrical and Electronics Engineers Inc. |
ISBN (Electronic) | 9798350323481 |
DOIs | |
State | Published - 2023 |
Externally published | Yes |
Event | 60th ACM/IEEE Design Automation Conference, DAC 2023 - San Francisco, United States Duration: Jul 9 2023 → Jul 13 2023 |
Publication series
Name | Proceedings - Design Automation Conference |
---|---|
Volume | 2023-July |
ISSN (Print) | 0738-100X |
Conference
Conference | 60th ACM/IEEE Design Automation Conference, DAC 2023 |
---|---|
Country/Territory | United States |
City | San Francisco |
Period | 7/9/23 → 7/13/23 |
Bibliographical note
Publisher Copyright:© 2023 IEEE.
Keywords
- Algorithm-Accelerator Co-Design
- Near-Memory Processing
- Neural Radiance Field
- On-Device Training