ACCURACY-AWARE MODEL COMPRESSION FOR REAL-TIME VISUAL PERCEPTION UNDER RESOURCE CONSTRAINTS
Journal of Computer Science and Electrical Engineering · 04.12.2022 · 4 (1) · 45-52
DOI: 10.61784/jcsee4154 ↗Kurz gefasst
Fachartikel in der Zeitschrift Journal of Computer Science and Electrical Engineering, online erschienen am 04.12.2022. Verfasst von FengYuan Zhang und 1 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0).
Abstract (Originalsprache)
Deploying real-time visual perception models on CPU-only edge devices requires compression that respects measured latency while preserving task accuracy. Uniform quantization and pruning ignore the uneven sensitivity of network components, and existing allocation methods optimize a single compression dimension with additive per-layer cost models that miss the precision-switch overheads of CPU inference runtimes. This paper presents ACORN, a post-training, accuracy-aware compression framework that jointly allocates channel-pruning ratios and numerical precision to the components of a pretrained detector under explicit latency and memory budgets. ACORN estimates the accuracy cost of every option as a perturbation of the uniformly quantized deployment model with a confidence-weighted discrepancy restricted to task-relevant categories, predicts latency from per-node profiles of the runtime-optimized graph augmented with pairwise quantize/dequantize costs, prunes channels without training through ridge least-squares reconstruction, and solves the allocation exactly as an integer linear program. On YOLOv8n deployed for PASCAL VOC detection on a single CPU core with ONNX Runtime, ACORN reaches 63.47% mAP@[.5:.95] at the latency of uniform INT8 quantization (62.77%), within 0.07 points of the FP32 model while running 2.28× faster with a 3.6× smaller model. At a 2.51× speedup it retains 63.17%, whereas latency-aware knapsack pruning and uniform DepGraph pruning at similar latency fall to 52.54% and 30.33%. The complete allocation uses 32 unlabeled images and about ten minutes of single-core computation.
Zitieren
APA
Zhang, F., & Yang, J. (2022). ACCURACY-AWARE MODEL COMPRESSION FOR REAL-TIME VISUAL PERCEPTION UNDER RESOURCE CONSTRAINTS. Journal of Computer Science and Electrical Engineering, 4(1), 45-52. https://doi.org/10.61784/jcsee4154
Vancouver
Zhang F, Yang J. ACCURACY-AWARE MODEL COMPRESSION FOR REAL-TIME VISUAL PERCEPTION UNDER RESOURCE CONSTRAINTS. Journal of Computer Science and Electrical Engineering. 2022;4(1):45-52. doi:10.61784/jcsee4154
BibTeX
@article{Zhang2022,
author = {Zhang, FengYuan and Yang, JieSi},
title = {ACCURACY-AWARE MODEL COMPRESSION FOR REAL-TIME VISUAL PERCEPTION UNDER RESOURCE CONSTRAINTS},
journal = {Journal of Computer Science and Electrical Engineering},
year = {2022},
volume = {4},
number = {1},
pages = {45--52},
doi = {10.61784/jcsee4154}
}
EUVIMED