DANMP: Accelerating Multi-Scale Deformable Attention Using Near-Memory-Processing Architecture
Open Access
- 5 July 2026
- conference paper
- Published by Association for Computing Machinery (ACM)
- p. 576-589
Abstract
No abstract availableAll Related Versions
This publication has 72 references indexed in Scilit:
- Accelerating Sparse Attention with a Reconfigurable Non-volatile Processing-In-Memory Architecture2023
- HAIMA: A Hybrid SRAM and DRAM Accelerator-in-Memory Architecture for Transformer2023
- DN-DETR: Accelerate DETR Training by Introducing Query DeNoising2022
- OW-DETR: Open-world Detection Transformer2022
- MAT: Processing In-Memory Acceleration for Long-Sequence Attention2021
- End-to-End Object Detection with Transformers2020
- DOTA: A Large-Scale Dataset for Object Detection in Aerial Images2018
- Understanding of Object Detection Based on CNN Family and YOLOJournal of Physics: Conference Series, 2018
- Microsoft COCO: Common Objects in Context2014
- The Pascal Visual Object Classes (VOC) ChallengeInternational Journal of Computer Vision, 2009