TreeFormer: a Semi-Supervised Transformer-based Framework for Tree Counting from a Single High Resolution Image

TL;DR

TreeFormer is a novel semi-supervised transformer-based framework that effectively estimates tree density and counts from single high-resolution images, reducing annotation costs and outperforming existing methods.

Contribution

Introduces the first semi-supervised transformer framework for tree counting, incorporating multi-scale features, a pyramid learning strategy, and a global tree count regulation mechanism.

Findings

Outperforms state-of-the-art semi-supervised methods.

Surpasses fully-supervised methods with the same labeled data.

Effective on multiple benchmark datasets.

Abstract

Automatic tree density estimation and counting using single aerial and satellite images is a challenging task in photogrammetry and remote sensing, yet has an important role in forest management. In this paper, we propose the first semisupervised transformer-based framework for tree counting which reduces the expensive tree annotations for remote sensing images. Our method, termed as TreeFormer, first develops a pyramid tree representation module based on transformer blocks to extract multi-scale features during the encoding stage. Contextual attention-based feature fusion and tree density regressor modules are further designed to utilize the robust features from the encoder to estimate tree density maps in the decoder. Moreover, we propose a pyramid learning strategy that includes local tree density consistency and local tree count ranking losses to utilize unlabeled images into the training process. Finally, the tree counter token is introduced to regulate the network by computing the global tree counts for both labeled and unlabeled images. Our model was evaluated on two benchmark tree counting datasets, Jiangsu, and Yosemite, as well as a new dataset, KCL-London, created by ourselves. Our TreeFormer outperforms the state of the art semi-supervised methods under the same setting and exceeds the fully-supervised methods using the same number of labeled images. The codes and datasets are available at https://github.com/HAAClassic/TreeFormer.