Efficient Spatio-Temporal Vegetation Pixel Classification with Vision Transformers

TL;DR

This paper optimizes Vision Transformers for efficient plant pixel classification over time, significantly reducing computational costs while maintaining accuracy in high-resolution vegetation monitoring.

Contribution

It provides a comprehensive ablation study on ViT design choices and demonstrates their effectiveness for scalable, resource-efficient spatio-temporal vegetation analysis.

Findings

ViT reduces FLOPs by an order of magnitude compared to CNNs.

ViT maintains constant parameter complexity regardless of time series length.

Experimental results show competitive classification performance on Brazilian Cerrado datasets.

Abstract

Plant phenology-the study of recurrent life cycle events-is essential for understanding ecosystem dynamics and their responses to climate change impacts. While Unmanned Aerial Vehicles (UAVs) and near-surface cameras enable high-resolution monitoring, identifying plant species across time remains computationally challenging. State-of-the-art approaches, specifically Multi-Temporal Convolutional Networks (CNNs), rely on rigid multi-branch architectures that scale poorly with longer time series and require large spatial context windows. In this paper, we present an extensive study on optimizing Vision Transformers (ViTs) for efficient spatio-temporal vegetation pixel classification. We conducted a comprehensive ablation study analyzing seven key design dimensions, including: (i) data normalization; (ii) spectral arrangement; (iii) boundary handling; (iv) spatial context window shape and size; (v) tokenization strategies; (vi) positional encoding; and (vii) feature aggregation strategies. Our method was evaluated on two datasets from the Brazilian Cerrado biome, Serra do Cip\'o (aerial imagery) and Itirapina (near-surface imagery). Experimental results demonstrate that our ViT approach offers a substantial improvement in computational efficiency while maintaining competitive classification performance. Notably, our ViT reduces Floating Point Operations (FLOPs) by an order of magnitude and maintains constant parameter complexity regardless of the time series length, whereas the CNN baseline scales linearly. Our findings confirm that ViTs are a robust, scalable solution for resource-constrained phenological monitoring systems.