Multi-Task Partially Supervised Learning for Super-Resolution and Semantic Segmentation on Earth Observation data

arXiv:2610.06389v1 Announce Type: new Abstract: Super-resolution and semantic segmentation are known to benefit one another, especially in the Earth observation context. However, learning both tasks in a joint model often requires both task annotations, which is impractical and expensive. In this paper, we study the multi-task partially supervised learning paradigm for both tasks, where each example is assumed to have only a single-task annotation. To that end, we examine two multi-task architectural variations, the sequential and shared variants, and then propose a hybrid variant and a re-projection loss to benefit from the shared representation and enforce image quality of super-resolution when training with semantic segmentation. Experiments show favorable results compared to the SOTA sequential variant. Source code will be published at https://github.com/lhoangan/munera.

Related

• Self-Supervised Super-Resolution for Sentinel-5P Hyperspectral Images
• UniverSat: Resolution- and Modality-Agnostic Transformers for Earth Observation
• FaithIR: Rethinking Infrared Image Super-Resolution from Perceptual Sharpness to Task Relevant Fidelity
• Multiclass Semantic Segmentation of Wildland Fire Images Using Context-Aware Centralized Copy-Paste Data Augmentation

Source: arXiv cs.CV | 2026-10-06