用视觉语言模型联合评估无人机图像整体、目标和背景质量。
DroneIQA-VLE: Multi-Task Drone Image Quality Assessment via Vision-Language Ensemble
- 融合视觉与语言双路径，分别用SigLIP2和微调的Qwen3.5-9B模型预测质量。
- 在低空无人机图像质量评估挑战中取得第二名，综合评分领先。
- 适合需要多维度图像质量分析的无人机视觉系统研发者。
我们提出DroneIQA-VLE，参加ICME 2026无人机图像质量评估挑战赛，旨在实现面向低空无人机图像的目标感知图像质量评估。该框架通过集成两种互补的处理流程，联合预测全局、目标和背景三个维度的质量分数：(1) 使用SigLIP2视觉编码器搭配多任务回归头；(2) 对Qwen3.5-9B多模态大语言模型进行LoRA适配后用于质量评分回归。最终的全局质量预测结果通过两路输出的算术平均获得。本方法在挑战赛中位列第二，验证了其有效性。代码已公开于https://github.com/sunwei925/DroneIQA-VLE。
原文摘要 · Abstract (English)
We present DroneIQA-VLE, our solution to the ICME 2026 Drone-IQA Grand Challenge on Target-aware Image Quality Assessment for Low-altitude UAV Images. The framework jointly predicts global, target, and background quality scores by ensembling two complementary pipelines: (1) SigLIP2 vision encoders with multi-task regression heads, and (2) a LoRA-adapted Qwen3.5-9B multimodal large language model for quality score regression. The final global quality prediction is obtained by arithmetically averaging the outputs of both pipelines. Our method achieves 2nd place in the challenge, demonstrating its effectiveness. The code is available at https://github.com/sunwei925/DroneIQA-VLE.
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。