EUVIMEDEUVIMED entdecken
Open Access · CC BY 4.0

MonoLENS: Monocular Lightweight Efficient Network with Separable Convolutions for Self-Supervised Monocular Depth Estimation

Genki Higashiuchi, Tomoyasu Shimada, Xiangbo Kong, Haimin Yan, Hiroyuki Tomiyama

Applied Sciences · 25.09.2025 · 15 (19) · 10393

DOI: 10.3390/app151910393 ↗

Kurz gefasst

Fachartikel in der Zeitschrift Applied Sciences, online erschienen am 25.09.2025. Verfasst von Genki Higashiuchi und 4 weiteren Autor:innen. Die Arbeit wurde bisher 6-mal zitiert (Crossref) und nennt 60 Literaturangaben. Der Volltext ist frei zugänglich (Open Access, CC BY 4.0). Gefördert von: Suzuki Foundation.

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

Self-supervised monocular depth estimation is gaining significant attention because it can learn depth from video without needing expensive ground-truth data. However, many self-supervised models remain too heavy for edge devices, and simply shrinking them tends to degrade accuracy. To address this trade-off, we present MonoLENS, an extension of Lite-Mono. MonoLENS follows a design that reduces computation while preserving geometric fidelity (relative depth relations, boundaries, and planar structures). MonoLENS advances Lite-Mono by suppressing computation on paths with low geometric contribution, focusing compute and attention on layers rich in structural cues, and pruning redundant operations in later stages. Our model incorporates two new modules, the DS-Upsampling Block and the MCACoder, along with a simplified encoder. Specifically, the DS-Upsampling Block uses depthwise separable convolutions throughout the decoder, which greatly lowers floating-point operations (FLOPs). Furthermore, the MCACoder applies Multidimensional Collaborative Attention (MCA) to the output of the second encoder stage, helping to make edge details sharper in high-resolution feature maps. Additionally, we simplified the encoder’s architecture by reducing the number of blocks in its fourth stage from 10 to 4, which resulted in a further reduction of model parameters. When tested on both the KITTI and Cityscapes benchmarks, MonoLENS achieved leading performance. On the KITTI benchmark, MonoLENS reduced the number of model parameters by 42% (1.8M) compared with Lite-Mono, while simultaneously improving the squared relative error by approximately 4.5%.

Förderung

  • Suzuki Foundation

Zitieren

APA

Higashiuchi, G., Shimada, T., Kong, X., Yan, H., & Tomiyama, H. (2025). MonoLENS: Monocular Lightweight Efficient Network with Separable Convolutions for Self-Supervised Monocular Depth Estimation. Applied Sciences, 15(19), 10393. https://doi.org/10.3390/app151910393

Vancouver

Higashiuchi G, Shimada T, Kong X, Yan H, Tomiyama H. MonoLENS: Monocular Lightweight Efficient Network with Separable Convolutions for Self-Supervised Monocular Depth Estimation. Applied Sciences. 2025;15(19):10393. doi:10.3390/app151910393

BibTeX
@article{Higashiuchi2025,
  author = {Higashiuchi, Genki and Shimada, Tomoyasu and Kong, Xiangbo and Yan, Haimin and Tomiyama, Hiroyuki},
  title = {MonoLENS: Monocular Lightweight Efficient Network with Separable Convolutions for Self-Supervised Monocular Depth Estimation},
  journal = {Applied Sciences},
  year = {2025},
  volume = {15},
  number = {19},
  pages = {10393},
  doi = {10.3390/app151910393}
}
Kostenlos starten →