用多位置嵌入提升小目标检测，参数量少效果好
Multi-Point Positional Insertion Tuning for Small Object Detection
- 在冻结模型中插入多个位置嵌入，增强特征定位能力
- 在SODA-D数据集上性能接近传统微调方法，参数量减少显著
- 适合资源有限时优化小目标检测模型
小目标检测旨在图像中定位和分类小型物体。随着大规模视觉-语言预训练的发展，微调预训练目标检测模型成为一种有前景的方法。然而，微调大型模型在计算和内存上成本高昂。为此，本文提出多点位置插入（MPI）微调，一种针对小目标检测的参数高效微调（PEFT）方法。具体而言，MPI将多个位置嵌入引入冻结的预训练模型，通过为潜在特征提供精确的位置信息，实现对小目标的高效检测。实验表明，该方法在SODA-D数据集上表现良好，性能与传统的PEFT方法（如CoOp和VPT）相当，同时显著减少了需微调的参数数量。
原文摘要 · Abstract (English)
Small object detection aims to localize and classify small objects within images. With recent advances in large-scale vision-language pretraining, finetuning pretrained object detection models has emerged as a promising approach. However, finetuning large models is computationally and memory expensive. To address this issue, this paper introduces multi-point positional insertion (MPI) tuning, a parameter-efficient finetuning (PEFT) method for small object detection. Specifically, MPI incorporates multiple positional embeddings into a frozen pretrained model, enabling the efficient detection of small objects by providing precise positional information to latent features. Through experiments, we demonstrated the effectiveness of the proposed method on the SODA-D dataset. MPI performed comparably to conventional PEFT methods, including CoOp and VPT, while significantly reducing the number of parameters that need to be tuned.
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。