Analyzing the Impact of Low-Rank Adaptation for Cross-Domain Few-Shot Object Detection in Aerial Images
This work addresses efficient adaptation for few-shot object detection in aerial images, but it is incremental as it applies an existing method to a new domain with modest gains.
This paper tackled the problem of cross-domain few-shot object detection in aerial images by applying Low-Rank Adaptation (LoRA) to small models, finding that LoRA slightly improves performance in low-shot settings like 1-shot and 5-shot, while full fine-tuning is more effective with more shots.
This paper investigates the application of Low-Rank Adaptation (LoRA) to small models for cross-domain few-shot object detection in aerial images. Originally designed for large-scale models, LoRA helps mitigate overfitting, making it a promising approach for resource-constrained settings. We integrate LoRA into DiffusionDet, and evaluate its performance on the DOTA and DIOR datasets. Our results show that LoRA applied after an initial fine-tuning slightly improves performance in low-shot settings (e.g., 1-shot and 5-shot), while full fine-tuning remains more effective in higher-shot configurations. These findings highlight LoRA's potential for efficient adaptation in aerial object detection, encouraging further research into parameter-efficient fine-tuning strategies for few-shot learning. Our code is available here: https://github.com/HichTala/LoRA-DiffusionDet.