# Counting Trees from Satellite Imagery with Noisy Supervision

arXiv: https://arxiv.org/abs/2606.24786 · HTML: https://mlsanity.com/paper/2606.24786/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 22 of its 53 references that are in the map.
- First posted on arXiv 23 Jun 2026.

## Builds on (most cited first, 22 of 22)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Sinkhorn Distances: Lightspeed Computation of Optimal Transportation Distances](https://mlsanity.com/paper/1306.0895/) (2013)
- [Objects as Points](https://mlsanity.com/paper/1904.07850/) (2019)
- [Bayesian Loss for Crowd Count Estimation with Point Supervision](https://mlsanity.com/paper/1908.03684/) (2019)
- [Rethinking Counting and Localization in Crowds:A Purely Point-Based Framework](https://mlsanity.com/paper/2107.12746/) (2021)
- [Distribution Matching for Crowd Counting](https://mlsanity.com/paper/2009.13077/) (2020)
- [A high-resolution canopy height model of the Earth](https://mlsanity.com/paper/2204.08322/) (2022)
- [Point in, Box out: Beyond Counting Persons in Crowds](https://mlsanity.com/paper/1904.01333/) (2019)
- [Very high resolution canopy height maps from RGB imagery using self-supervised vision transformer and convolutional decoder trained on Aerial Lidar](https://mlsanity.com/paper/2304.07213/) (2023)
- [Iterative Crowd Counting](https://mlsanity.com/paper/1807.09959/) (2018)
- [Counting with Focus for Free](https://mlsanity.com/paper/1903.12206/) (2019)
- [Estimating Canopy Height at Scale](https://mlsanity.com/paper/2406.01076/) (2024)
- [Open-Canopy: Towards Very High Resolution Forest Monitoring](https://mlsanity.com/paper/2407.09392/) (2024)
- [OAM-TCD: A globally diverse dataset of high-resolution tree cover maps](https://mlsanity.com/paper/2407.11743/) (2024)
- [Individual Tree Detection in Large-Scale Urban Environments using High-Resolution Multispectral Imagery](https://mlsanity.com/paper/2208.10607/) (2022)
- [PureForest: A Large-Scale Aerial Lidar and Aerial Imagery Dataset for Tree Species Classification in Monospecific Forests](https://mlsanity.com/paper/2404.12064/) (2024)
- [TreeFormer: a Semi-Supervised Transformer-based Framework for Tree Counting from a Single High Resolution Image](https://mlsanity.com/paper/2307.06118/) (2023)
- [Get Your Embedding Space in Order: Domain-Adaptive Regression for Forest Monitoring](https://mlsanity.com/paper/2405.00514/) (2024)
- [Bridging Classical and Modern Computer Vision: PerceptiveNet for Tree Crown Semantic Segmentation](https://mlsanity.com/paper/2505.23597/) (2025)
- [MAP-UOT: A Memory-Efficient Approach to Unbalanced Optimal Transport Implementation](https://mlsanity.com/paper/2412.11079/) (2024)
- [Trees as Gaussians: Large-Scale Individual Tree Mapping](https://mlsanity.com/paper/2508.21437/) (2025)

## Abstract (arXiv)

Counting individual trees is a fundamental task for environmental monitoring, yet remains largely unexplored with satellite imagery. At these resolutions, isolated trees may still be identifiable, but crown boundaries become ambiguous in dense forests, making the notion of an individual tree inherently ill-defined. Moreover, large-scale manual annotations of individual trees are prohibitively expensive. While scalable supervision can be derived from airborne LiDAR, the resulting annotations are noisy and difficult to exploit effectively. We address these challenges by formulating tree counting as a spatial density matching problem supervised through Unbalanced Optimal Transport. This formulation naturally accommodates both precise localization of isolate trees and robust density estimation in dense forests. We further introduce a self-correction mechanism that leverages transport residuals to progressively refine noisy supervision during training. We evaluate our approach on TinyTrees, a new benchmark spanning three continents and three satellite sensors, comprising over 216 million tree annotations (including 639k manually verified instances) across 25,890 km2. Our method consistently outperforms detection-based, regression-based, and transport-based distribution-matching baselines, demonstrating the effectiveness of unbalanced transport and reliability-aware supervision for large-scale tree counting from satellite imagery. Code, data and models are available at https://github.com/dgominski/treematch.