Functional Map of the World Challenge

Can you build algorithms to classify facility, building, and land use from satellite imagery?

Registration begins July 2017

The Challenge

Create fast and accurate building and land use classification algorithms

How To Enter

Register now at


Compete for $100,000 in prizes

The Problem

Intelligence analysts, policy makers, and first responders around the world rely on geospatial land use data to inform crucial decisions about global defense and humanitarian activities. Historically, analysts have manually identified and classified geospatial information by comparing and analyzing satellite images, but that process is time consuming and insufficient to support disaster response.

The Functional Map of the World (fMoW) Challenge seeks to foster breakthroughs in the automated analysis of overhead imagery by harnessing the collective power of the global data science and machine learning communities. The challenge will publish one of the largest publicly available satellite-image datasets to date, with more than one million points of interest from around the world. The dataset contains satellite-specific metadata that researchers can exploit to build a competitive algorithm that classifies facility, building, and land use.

Be Part of the Innovation

IARPA is conducting this Challenge to invite the broader research community of industry and academia, with or without experience in deep learning and computer vision analysis, to participate in a convenient, efficient and non-contractual way. Participants will develop algorithms that scan satellite data to identify functions based on multiple reference sources, such as overhead and ground-based images, digital elevation data, existing well-understood image collections, surface geology, geography, and cultural information. The goals and objectives of this Challenge are to:

Challenge Details
Begins Jul 13, 2017

Participants will register for the challenge at

Datasets & Training
Jul & Aug 2017

IARPA will release a point-of-interest classification library and image sets as training data through an AWS Public Data Set. Released in two batches, the library will include 62 pre-defined categories, and the final image set will contain 1,000,000 “chips,” each with a bounding box around an unidentified point of interest. Participants will adjust their neural networks and develop algorithms based on the data, and a visualizer will be provided to test the solutions. A pre-trained model and baseline will be available to start the training process.

The data is available via Requestor Pays in two versions:

  • RGB JPG Data Set: arn:aws:s3:::fmow-rgb | s3://fmow-rgb
  • Multispectral TIFF Data Set: arn:aws:s3:::fmow-full | s3://fmow-full

A full set of AWS CLI resources can be found here:
Some example commands appear below:

There is a manifest.json.bz2 file in each bucket that can be downloaded to get a json that lists everyfile in the bucket

  • aws s3api get-object --bucket fmow-rgb --key manifest.json.bz2 --request-payer requester
  • aws s3api get-object --bucket fmow-full --key manifest.json.bz2 --request-payer requester

Commands like these can be used to get a directory listing

  • aws s3 ls s3://fmow-rgb --request-payer requester
  • aws s3 ls s3://fmow-full --request-payer requester

Provisional Scoring
Aug 2017

The Provisional scoring mechanism and leaderboard will be released. Provisional scores will be posted to the challenge leaderboard, and participants will have the opportunity to retune their algorithms to increase accuracy. All participants will have access to IBM Watson and IBM Bluemix for 90 days during the challenge (though these are not required and we welcome all types of solutions for the challenge). The top 10 participants will have access to AWS cloud computing resources for 10 days during this phase.

Final Submission
Dec 2017

The challenge submission period will end. The final score shown on the Provisional leaderboard at the end of the challenge will be used to determine solver rankings going into the final evaluation. The top 10 algorithms will be scored against a hidden data set, and the top scoring solutions will be validated by the IARPA team for award. Final scores will be posted to the leaderboard on Topcoder and shared through official IARPA communications.

Awards & Recognition
Feb 2018

The challenge winners will be invited to present their solutions to IARPA and other key leaders in the Government at a workshop in Washington, DC, and cash awards will be distributed to winners. Final public communication about winners will take place during this day-long workshop.


Participants will be eligible to win cash prizes from a total prize purse of $100,000. Additionally, top winners will get a chance to present their winning solutions at a workshop in Washington, D.C. Prizes will be distributed for the following criteria:

First Prize


Second Prize


Third Prize


Fourth Prize


Fifth Prize


Best POI Category


Best Undergraduate


Milestone Awards (each month)

$9,000 (total)

Open Source Incentives (x3)

$15,000 (total)

Rules & Eligibility

Anyone 18 years or older is eligible to register. Certain individuals and groups with existing agreements with IARPA, IARPA government partners, and their affiliates are welcome to participate in the challenge, but will need to forego the monetary prizes, but may compete for standing on the leaderboard and other non-monetary incentives. Participants can also form teams to collaborate on solutions. More information about rules and eligibility coming soon.


To assist builders in this challenge, IARPA has gathered a list of resources to prepare for each stage in the challenge. You can find sample datasets, reference capture methods, evaluation guides, and technical documents to support your submissions.

Image-Net Database

Large repository of images with associated bounding boxes
COCO Dataset

Image recognition, segmentation, and captioning dataset.

Standardized image data sets for object class recognition
Kaggle Competition Example

Satellite data competition: Understanding the Amazon from Space
Kaggle Dstl Satellite Imagery Feature Detection Challenge

Challenge from the Defence Science and Technology Laboratory (Dstl) to accurately classify features in overhead imagery
Land Use Dataset

UC Merced image dataset with 21 classes with 100 images per class

SpaceNet online repository of freely available satellite imagery

Publicly available benchmark for Remote Sensing Image Scene Classification (RESISC), created by Northwestern Polytechnical University (NWPU)
AID Project

Benchmark Dataset for Performance Evaluation of Aerial Scene Classification

Google image dataset of mainly urban areas of China
TorontoCity: Seeing the World with a Million Eyes

TorontoCity benchmark providing different perspectives of the world captured from airplanes, drones and cars driving around the city
Remote Sensing Image Scene Classification: Benchmark and State of the Art

Comprehensive review of the recent progress in dataset progress. Propose a large-scale dataset, termed "NWPU-RESISC45"
TensorFlow Tutorial

This guide gets you started programming in TensorFlow to help build you neural net
The Neural Network Zoo

A cheat sheet guide containing many neural network architectures and their explanations
Amazon AI Services

A listing of Amazon AI Services that can be used on the challenge
fMoW Benchmark Example

Benchmark Algorithm and code used for the fMoW Challenge

Coming Soon
SpaceNet GitHub Repository

Packages intended to assist in the preprocessing of SpaceNet satellite imagery data corpus to a format that is consumable by machine learning algorithms.
DigitalGlobe DeepCore SDK

DeepCore Machine Learning Abstraction Framework is a utility toolkit that allows a user to download, perform either image classification or object detection, and manipulate geospatial vector files.
Connect With Us

Learn more at

By Email

Reach out directly at

On Twitter

Find us on Twitter at @IARPAnews & #IARPAfMoW


The Intelligence Advanced Research Projects Activity (IARPA) invests in high-risk, high-payoff research programs to tackle some of the most difficult challenges of the agencies and disciplines in the Intelligence Community (IC)