Pose Bowl: Pose Estimation Track

Develop algorithms for use on inspector spacecraft that take and process photos of other ships. In the Pose Estimation Track, solutions will identify the position and orientation of the spacecraft's camera across sequences of images. #science

$28,000 in prizes
4 days left
401 joined

Code submission format

This is a code submission challenge!

Rather than submitting your predicted labels, you'll package everything needed to perform inference and submit that for containerized execution. The runtime repository contains the complete specification for the runtime.

What to submit


Your submission will be a zip archive (e.g. submission.zip). The root level of the submission.zip file must contain a main.sh shell script which performs inference on all of the test images in the data/ directory of the code execution environment and writes your predictions into a CSV file in the submission/ directory.

Here's an example of what a submission containing a simple script that loads a pre-trained model might look like. Note that in this example, the main.sh could call a python script main.py but you are not required to use python or even have an additional script. The only file that must exist in your submission.zip is main.sh.

submission.zip
├── assets           # Configuration and serialized model file
│   ├── model.pt
│   └── settings.yml
├── main.sh          # Required: Shell script called by code execution entrypoint
└── main.py          # Python script you provide called by main.sh

To help develop and test your solution, we provide you with the runtime repository, which contains the following:

  • An example main.sh script and a subset of the challenge data to help get you started.
  • A scoring script scoring/score_submission.py for calculating the challenge metric which you can use locally to evaluate your predictions.
  • A complete specification for the code execution runtime.

Note that the runtime environment has many standard Python libraries for computer vision already installed, but if you would like to, you may also add Python or other dependencies to the environment by following the process described in "Updating runtime packages" section of the runtime README.

Runtime constraints


Your code will be executed within a container that is defined in our runtime repository. The limits are as follows:

  • Your submission must use the packages defined in the runtime repository. See the README of the runtime repo for details on how to request additional packages.
  • Your submission must complete execution in 2.5 hours or less.
  • The container has access to 3 cores and 4 GB of RAM. There is no GPU.
  • The container will not have network access. All necessary files (code and model assets) must be included in your submission.
  • The container execution will not have root access to the filesystem.

Please also be aware that the machines for executing your code are a shared resource across competitors, so please be conscientious in your use of them. You are limited to in the number of submissions you can make (though canceled or failed jobs won't count against your daily submission limit). Thoroughly test your submissions locally, add progress information to your logs, and cancel jobs that will run longer than the time limit.