SyMTRS: Benchmark Multi-Task Synthetic Dataset for Depth, Domain Adaptation and Super-Resolution in Aerial Imagery

2026-04-23 · Source: Computer Vision and Pattern Recognition · Field: Technology & Digital — Artificial Intelligence & Machine Learning, Robotics & Autonomous Systems · Depth: Expert, quick

Summary

SyMTRS is a new large-scale synthetic dataset designed to address critical gaps in remote sensing research, particularly for aerial imagery tasks. Generated via a high-fidelity urban simulation pipeline, it provides high-resolution RGB aerial images (2048 x 2048), pixel-perfect depth maps, and night-time counterparts for domain adaptation. The dataset also includes aligned low-resolution variants for super-resolution at x2, x4, and x8 scales. Unlike existing remote sensing datasets that typically focus on a single task or modality, SyMTRS is a unified multi-task benchmark. It enables joint research in monocular depth estimation, cross-domain robustness, and resolution enhancement, offering controlled experiments with perfect geometric ground truth and consistent multi-domain supervision.

Key takeaway

For research scientists developing deep learning models for remote sensing, SyMTRS offers a unique opportunity to overcome data scarcity for complex tasks. You should consider integrating SyMTRS into your experimental pipelines to validate models for monocular depth estimation, domain adaptation, and super-resolution, leveraging its perfect ground truth and multi-task design to accelerate development and improve robustness.

Key insights

SyMTRS is a multi-task synthetic dataset for aerial imagery, providing perfect ground truth for depth, domain adaptation, and super-resolution.

Principles

• Synthetic data can bridge annotation gaps.
• Multi-task datasets enable joint research.
• High-fidelity simulation yields pixel-perfect ground truth.

Method

The dataset is generated using a high-fidelity urban simulation pipeline to produce RGB, depth, night-time, and multi-scale low-resolution aerial imagery.

In practice

• Use SyMTRS for monocular depth estimation.
• Apply SyMTRS to domain adaptation tasks.
• Test super-resolution models with SyMTRS.

Topics

• SyMTRS
• Synthetic Dataset
• Aerial Imagery
• Monocular Depth Estimation
• Domain Adaptation

Code references

• safouaneelg/SyMTRS

Best for: Research Scientist, AI Scientist, Machine Learning Engineer, Computer Vision Engineer

Related on AIssential

• FusionRS: A Large-Scale RGB-Infrared Remote Sensing Dataset for Dual-Modal Vision-Language Foundation Models — FusionRS enables dual-modal RGB-infrared vision-language models through a novel dataset with IR-aware captions.
• Researchers are combining drones and AI to make removing land mines faster and safer — Drone-based multisensor AI improves land mine detection speed, accuracy, and safety by fusing diverse data and quantifying uncertainty.
• SADGE: Structure and Appearance Domain Gap Estimation of Synthetic and Real Data — Synthetic data utility for computer vision tasks is best predicted by the non-linear interplay of appearance and geometric similarities, not either alone.
• MapSR: Prompt-Driven Land Cover Map Super-Resolution via Vision Foundation Models — MapSR uses prompt-driven vision foundation models for efficient, training-free land-cover super-resolution with minimal annotation.
• 🪝Syn4D: Multiview Synthetic 4D Dataset🪝 👉Syn4D is novel multi-view synthetic dataset of... — Syn4D is a synthetic multi-view 4D dataset offering comprehensive ground-truth for dynamic scene analysis.
• Honey, I Shrunk the Arc de Triomphe! — A new dataset and method mitigate scale-collapse in monocular depth estimation for distant, unconstrained scenes.

Open in AIssential →

Editorial summary, takeaway, and curation by AIssential. Original article published by Computer Vision and Pattern Recognition.