轻量级检测器RF-DETR通过神经架构搜索实现实时高精度检测。
RF-DETR: Neural Architecture Search for Real-Time Detection Transformers
- 基于权重共享NAS，在不重训练情况下快速评估数千种配置
- RF-DETR(nano)在COCO上达48.0 AP，比D-FINE(nano)高5.3 AP
- 2x-large版在Roboflow100-VL上快20倍且精度超GroundingDINO(tiny)
开放词汇检测器在COCO上表现优异，但在包含分布外类的现实数据集上泛化能力差。本文提出RF-DETR，一种轻量级专用检测变压器，通过权重共享神经架构搜索（NAS）为任意目标数据集发现准确率-延迟帕累托曲线。该方法在目标数据集上微调预训练基础网络，并评估数千种不同准确率-延迟权衡的网络配置，无需重新训练。此外，重新审视NAS中的可调参数以提升DETR在多样化目标域的迁移能力。实验表明，RF-DETR显著优于现有实时检测方法：RF-DETR(nano)在COCO上达到48.0 AP，较D-FINE(nano)提升5.3 AP；RF-DETR(2x-large)在Roboflow100-VL上超越GroundingDINO(tiny) 1.2 AP，同时运行速度提升20倍。据我们所知，RF-DETR(2x-large)是首个在COCO上超过60 AP的实时检测器。
原文摘要 · Abstract (English)
Open-vocabulary detectors achieve impressive performance on COCO, but often fail to generalize to real-world datasets with out-of-distribution classes not typically found in their pre-training. Rather than simply fine-tuning a heavy-weight vision-language model (VLM) for new domains, we introduce RF-DETR, a light-weight specialist detection transformer that discovers accuracy-latency Pareto curves for any target dataset with weight-sharing neural architecture search (NAS). Our approach fine-tunes a pre-trained base network on a target dataset and evaluates thousands of network configurations with different accuracy-latency tradeoffs without re-training. Further, we revisit the "tunable knobs" for NAS to improve the transferability of DETRs to diverse target domains. Notably, RF-DETR significantly improves over prior state-of-the-art real-time methods on COCO and Roboflow100-VL. RF-DETR (nano) achieves 48.0 AP on COCO, beating D-FINE (nano) by 5.3 AP at similar latency, and RF-DETR (2x-large) outperforms GroundingDINO (tiny) by 1.2 AP on Roboflow100-VL while running 20x as fast. To the best of our knowledge, RF-DETR (2x-large) is the first real-time detector to surpass 60 AP on COCO. Our code is available at https://github.com/roboflow/rf-detr
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。