Parameter-Efficient Domain Adaptation and Lightweight Decoding for Agricultural Monocular Depth Estimation
Agronomy · 13.05.2026 · 16 (10) · 972
DOI: 10.3390/agronomy16100972 ↗Kurz gefasst
Fachartikel in der Zeitschrift Agronomy, online erschienen am 13.05.2026. Verfasst von Yanliang Mao und 2 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref) und nennt 26 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Regional Innovation Guidance Plan of the Science and Technology Bureau of Xinjiang Production and Construction Corps.
Abstract (Originalsprache)
Reliable monocular depth estimation (MDE) is essential for agricultural robots and unmanned platforms, where low-cost visual perception is required for safe navigation and scene understanding in complex field environments. However, general-purpose depth foundation models remain limited by substantial domain gaps in agriculture, while full fine-tuning of large backbones is computationally expensive and less suitable for deployment on resource-constrained platforms. In this paper, an efficient agricultural MDE framework, termed AgriLoRA-DA, is proposed based on Depth-Anything-V2. Specifically, the pretrained DINOv2 encoder is kept frozen and adapted using LoRA in selected attention projections, while the original Dense Prediction Transformer (DPT) decoder is replaced with a lightweight Lite-FPNHead to reduce decoding overhead and improve deployment efficiency. Experiments conducted on the WE3DS dataset indicate that, although Depth-Anything-V3 provides the strongest zero-shot generalization among the evaluated baselines, target-domain adaptation is still necessary for WE3DS agricultural scenes. After adaptation, AgriLoRA-DA achieves the best overall performance with AbsRel = 0.0133, SqRel = 3.518, RMSE = 132.264, log10 = 0.0057, and delta1 = 0.9990, while requiring only 0.19 M (0.87%) trainable parameters. These results suggest that parameter-efficient adaptation and lightweight decoding provide a practical direction for deployable depth estimation in crop-row scenes similar to WE3DS, while broader cross-dataset validation remains an important direction for future work.
Förderung
- Regional Innovation Guidance Plan of the Science and Technology Bureau of Xinjiang Production and Construction Corps
Zitieren
APA
Mao, Y., Zhao, W., & Chen, L. (2026). Parameter-Efficient Domain Adaptation and Lightweight Decoding for Agricultural Monocular Depth Estimation. Agronomy, 16(10), 972. https://doi.org/10.3390/agronomy16100972
Vancouver
Mao Y, Zhao W, Chen L. Parameter-Efficient Domain Adaptation and Lightweight Decoding for Agricultural Monocular Depth Estimation. Agronomy. 2026;16(10):972. doi:10.3390/agronomy16100972
BibTeX
@article{Mao2026,
author = {Mao, Yanliang and Zhao, Wenhao and Chen, Liping},
title = {Parameter-Efficient Domain Adaptation and Lightweight Decoding for Agricultural Monocular Depth Estimation},
journal = {Agronomy},
year = {2026},
volume = {16},
number = {10},
pages = {972},
doi = {10.3390/agronomy16100972}
}
EUVIMED