Computer Vision for 2D Data Analysis

National Oceanography Centre - Southampton

September 23-24, 2025

9:30 am - 5:00 pm BST (optionally 9:30 am - 12:00 noon BST on 25th)

Instructors: Eric Orenstein

Helpers: Alex Baldwin, Jonathan Coney, Esther Turner, Colin Sauze, Tobias Ferreira, Adam Ward

Some adblockers block the registration window. If you do not see the registration box below, please go directly to the registration page.

General Information

About this workshop: Ocean Informatics is organizing a two day workshop on September 23 & 24 that will teach computer vision and machine learning methods to NOC scientists, based on the CV4Ecology Short Course and POGO ML for Ocean Observations. Computer vision (CV) is significantly accelerating environmental research by automating the analysis of raw data from a variety of sources (e.g. camera traps, drones, and satellites). The goal of our short intro to CV is to expose NOC scientists to the software engineering, machine learning and CV skills necessary to analyse big sets of visual data on their own. We will focus on ecological image data as an example task, but the techniques introduced are quite generic. Participants working with images, video, acoustics, satellite imagery, and model output will gain basic skills and vocabulary to jump start their work.

Specifically, the workshop will include:

  • A short, high-level introduction on what CV is and how it can be used. We will avoid in-depth theoretical treatments and focus on applying existing tools in a principled fashion.
  • A hands-on project that will require participants to write their own Python code. We will supply a small, annotated image dataset in the domain of AI for wildlife conservation and enough code scaffolding so participants do not have to start from scratch. NB: Python experience is not necessary! Depending on interest, we will divide the workshop into small subgroups based on level of comfort with the language.
  • A maximum 5:1 student to teacher ratio to ensure everyone is able to get the help they need on the project or discuss general questions about machine learning.
  • Cloud-based computing resources for training and running models. Depending on interest, the compute will be provisioned either on the Data Science Platform or on JASMIN.
  • A follow-up workshop several weeks after the initial session to discuss progress on independent projects (i.e. data prep, model training, compute resources, etc).

    • We expect to run the course from 0930 to 1700 on the 23rd and 24th with a two-hour lunch break from 1200-1400. Instructors will additionally be available until 1200 on the 25th before the open staff meeting to help participants finish their work or prepare to deploy model training routines on JASMIN. Ideally, participants will clear their schedule or move critical meetings to the long lunch breaks for the two full days of the workshop. To get the most out of the time, everyone should plan to focus on the course as much as possible.

    If you are interested in participating, please register by close of play on the 11th September. If there are no tickets left, please get in contact.

    Check out ct_classifier repo and POGO imaging course to get a sense of the material we will cover.

    Where: Seminar Room, Library Red Room and Library Blue Room, National Oceanography Centre, European Way, Southampton, SO14 3ZH. Get directions with OpenStreetMap or Google Maps. What3Words location: ///super.neon.light.

    When: September 23-24, 2025; 9:30 am - 5:00 pm BST (optionally 9:30 am - 12:00 noon BST on 25th) Add to your Google Calendar.

    Requirements: Participants must bring a laptop which can connect to the Eduroam or NOCstaff wifi.

    Contact: Please email oi@noc.ac.uk for more information.

    Accessibility: We are committed to making this workshop accessible to everybody.

    We are dedicated to providing a positive and accessible learning environment for all. We do not require participants to provide documentation of disabilities or disclose any unnecessary personal information. However, we do want to help create an inclusive, accessible experience for all participants. We encourage you to share any information that would be helpful to make your Carpentries experience accessible.

    Code of Conduct

    Everyone who participates in Carpentries activities is required to conform to the Code of Conduct. This document also outlines how to report an incident if needed.

    Surveys

    Please be sure to complete these surveys before and after the workshop.

    Pre-workshop Survey

    Post-workshop Survey

    Schedule

    The lesson taught in this workshop is being piloted and a precise schedule is yet to be established. The workshop will include regular breaks, including a lunch break between noon and 2pm each day. Please contact the workshop organisers if you would like more information about the planned schedule.

    Setup

    To participate in a workshop, you will need access an up-to-date web browser.