---
title: "开箱即用 vs 微调：面向军事目标检测识别的新基准测试"
source: "arXiv AI/CL/LG（论文）"
source_tier: "t2"
url: "https://arxiv.org/abs/2608.17917v1"
permalink: "https://feed.astralor.com/items/itm_8110d1c5329d0adeb3d58528"
published: "2026-08-18T15:43:34.000Z"
score: 72
channel: "tech"
tags: ["DETR","VisDrone","YOLO","军事AI","目标检测"]
---

# 开箱即用 vs 微调：面向军事目标检测识别的新基准测试

## AI 点评

> 这篇论文的价值在于把注意力拉回‘数据和场景决定一切’这个朴素事实。对做军事视觉或边缘部署的团队来说，开箱即用模型的表现和跨域微调的实际收益都是预算分配时最需要的数据点。即便你不在军工领域，A2G 小目标检测的困境也值得所有做无人机视觉的人警醒。

## 速览

本文针对军事自动目标检测与识别（ATD/R）场景，评估了 YOLO 系列六代模型与 DETR 框架两种变体在开箱即用及 VisDrone 数据集微调后的性能。研究使用新建的军事相关数据集，涵盖军用车辆、遮挡和小目标等挑战，以 mAP@0.5 和 mAP@0.5:0.95 为指标，从空对地（A2G）与地对地（G2G）视角、目标尺寸和模型大小等维度进行比较。结果表明：大模型优于小模型，DETR 系列表现有潜力，跨域微调能提升 A2G 性能，但所有模型在 A2G 小目标检测上仍存在明显短板。结论认为，当前通用检测模型虽不断进步，领域内数据训练仍是构建有效 ATD/R 系统的关键。

## 关键点

- 首次在新建军事数据集上系统对比 YOLO 系列与 DETR 变体在开箱即用和微调下的性能。
- 大模型普遍优于小模型，DETR 在精度上展现出比 YOLO 系列更有潜力的表现。
- 基于 VisDrone 的跨域微调可提升空对地和小目标检测性能，但提升有限。
- 所有模型在空对地场景下的小目标检测仍然困难，领域内训练不可或缺。

## 全文

Automatic Target Detection and Recognition (ATD/R) is critical for military decision support and (semi-)autonomous operations. Recent advances in object detection and artificial intelligence (AI) significantly boosted the potential performance of ATD/R. However, the scarcity of publicly available military datasets limits the application of these systems. As a solution, this paper explores the use of publicly available models and civilian datasets to achieve reasonable performance in military contexts. We benchmark several state-of-the-art models, including six iterations of the YOLO series and two variations on the DETR framework, on a newly acquired military relevant dataset. This dataset features military vehicles and challenging circumstances, including various degrees of occlusions and small targets. The out-of-the-box version of each model is validated alongside a version finetuned on the VisDrone dataset. This dataset features small objects, an Air-to-Ground (A2G) perspective and relevant classes, potentially generalizing to our military ATD/R task. We compare the performance of the models using mAP@0.5 and mAP@0.5:0.95, across A2G and Ground-to-Ground (G2G) perspective, target size and model size, giving insight into the real-time capabilities of models. Our main findings are: (1) bigger models outperform smaller models, (2) DETR-based models show promising results compared to the YOLO series,(3) fine-tuning models on an out-of-domain A2G dataset, improves their A2G performance and slightly improves their performance on small objects, but (4) all models still struggle with detecting small objects in an A2G scenario. We conclude that, despite recent advances in object detection, in-domain training is still crucial for creating capable ATD/R systems.

---

引用前请核对原文;本文由星际流动整理:https://feed.astralor.com/items/itm_8110d1c5329d0adeb3d58528