Co-SemDepth: Fast Joint Semantic Segmentation and Depth Estimation on Aerial Images

Yara AlaaEldin, Francesca Odone
3/23/2025
cs.CVcs.AI

Official code: Malga-Vision/Co-SemDepth (11 stars, AGPL-3.0, last commit 2026-03-26).

Abstract

Understanding the geometric and semantic properties of the scene is crucial in autonomous navigation and particularly challenging in the case of Unmanned Aerial Vehicle (UAV) navigation. Such information may be by obtained by estimating depth and semantic segmentation maps of the surrounding environment and for their practical use in autonomous navigation, the procedure must be performed as close to real-time as possible. In this paper, we leverage monocular cameras on aerial robots to predict depth and semantic maps in low-altitude unstructured environments. We propose a joint deep-learning architecture that can perform the two tasks accurately and rapidly, and validate its effectiveness on MidAir and Aeroscapes benchmark datasets. Our joint-architecture proves to be competitive or superior to the other single and joint architecture methods while performing its task fast predicting 20.2 FPS on a single NVIDIA quadro p5000 GPU and it has a low memory footprint. All codes for training and prediction can be found on this link: https://github.com/Malga-Vision/Co-SemDepth

View on arXivView PDF

Code Implementations(1)

Malga-Vision/Co-SemDepthOfficial100%
AGPL-3.0

Official implementation for the joint network presented in the paper "Co-SemDepth: Fast Joint Semantic Segmentation and Depth Estimation on Aerial Images"

111C++, ShellNov 11, 20246 months agoAGPL-3.0

Similar Papers

Real-Time Monocular Scene Analysis for UAV in Outdoor Environments

92% match2026

Self-Supervised Learning to Fly using Efficient Semantic Segmentation and Metric Depth Estimation for Low-Cost Autonomous UAVs

84% match2025

NeoARCADE: Robust Calibration for Distance Estimation to Support Assistive Drones for the Visually Impaired

84% match2025

RoadRunner -- Learning Traversability Estimation for Autonomous Off-road Driving

84% match2024

Aerial Vision-Language Navigation with a Unified Framework for Spatial, Temporal and Embodied Reasoning

84% match2025

Vision-based Geo-Localization of Future Mars Rotorcraft in Challenging Illumination Conditions

84% match2025

Cite this paper

@article{alaaeldin2025cosemdepth,
  title  = {Co-SemDepth: Fast Joint Semantic Segmentation and Depth Estimation on Aerial Images},
  author = {Yara AlaaEldin and Francesca Odone},
  year   = {2025},
  eprint = {2503.17982},
  archivePrefix = {arXiv},
  url    = {http://arxiv.org/abs/2503.17982v1}
}

Add a badge to your README

Shows whether official code is linked for this paper.

Code of Paper badge
[![Code of Paper](https://codeofpaper.com/badge/2503.17982.svg)](https://codeofpaper.com/paper/2503.17982)

Discussion