Leveraging Image Generators to Address Training Data Scarcity: The Gen4Regen Dataset for Forest Regeneration Mapping

2026-05-06

AI Summary

This study addresses the challenges of scarce high-quality annotations and severe class imbalance in semantic segmentation of forest regeneration tree species by proposing a novel approach that integrates real aerial imagery with AI-generated images. Leveraging the large-scale vision-language model Nano Banana Pro, the method enables prompt-driven, simultaneous generation of high-fidelity synthetic images and their corresponding pixel-level semantic masks. The resulting Gen4Regen dataset, combined with the real-world WilDReF-Q-V2 dataset for joint training, significantly enhances fine-grained forest regeneration mapping performance. The approach achieves an overall F1-score improvement of over 15 percentage points, with gains of up to 30 percentage points for certain rare tree species, thereby effectively mitigating the critical bottleneck of limited annotated data in ecological remote sensing.

Abstract

Sustainable forest management relies on precise species composition mapping, yet traditional ground surveys are labour-intensive and geographically constrained. While Uncrewed Aerial Vehicles (UAVs) offer scalable data collection, the transition to deep learning-based interpretation is bottlenecked by the severe scarcity of expert-annotated imagery, particularly in complex, visually heterogeneous regeneration zones. This paper addresses the dual challenges of data scarcity and extreme class imbalance in the semantic segmentation of fine-grained forest regeneration species by providing a scalable framework that reduces reliance on manual photo-interpretation for high-resolution, millimetre-level aerial imagery. Importantly, we leverage the large-scale vision-language Nano Banana Pro model to simultaneously generate high-fidelity images and their corresponding pixel-aligned semantic masks from prompts. We introduce WilDReF-Q-V2, an expansion of a natural forest dataset with 13 977 new unlabelled and 50 labelled real images, as well as the Gen4Regen dataset, featuring 2101 pairs of synthetic images and semantic masks. Our methodology integrates real-world data with AI-generated images, highlighting that AI-generated data is highly complementary to real-world data, with unified training yielding an F1 score improvement of over 15 %pt compared to purely supervised baselines. Furthermore, we demonstrate that even small quantities of prompt-generated data significantly improve performance for underrepresented species, some of which saw per-species F1 score gains of up to 30 %pt. We conclude that vision-language models can serve as agile data generators, effectively bootstrapping perception tasks for niche AI domains where expert labels are scarce or unavailable. Our datasets, source code, and models will be available at https://norlab-ulaval.github.io/gen4regen.

Problem

• data scarcity
• class imbalance
• semantic segmentation
• forest regeneration
• expert-annotated imagery

Innovation

• synthetic data generation
• vision-language model
• semantic segmentation
• forest regeneration mapping
• data scarcity