Computer Vision for Ecology in South Africa

I. About

Making computer vision more accessible for ecology.

Camera traps, drones, underwater cameras and acoustic recorders are generating more ecological data than researchers can realistically analyse by hand. Computer vision (CV) and machine learning offer powerful ways to transform these vast datasets into meaningful insights about wildlife and ecosystems. Yet, knowing where to start and how to build reliable models remains a challenge for many researchers.

Computer Vision for Ecology at AIMS is a five-day, in-person workshop for ecologists, conservationists, students and researchers interested in using AI to tackle real-world ecological questions.

Through practical sessions and hands-on project work, participants will learn to build CV workflows from the ground up. Working with real ecological datasets, they will explore data preparation, model training and evaluation, investigate model performance, and experiment with ways to improve results.

Whether you work with camera-trap images, wildlife videos or acoustic recordings, this workshop will help you discover how CV can support your research and open new possibilities for analysing ecological data at scale.

Workshop format

Five days of practical learning, discussion and collaboration.

The workshop will combine lectures, practical sessions and ecological case studies, with a focus on applying computer vision to real-world research and conservation challenges.

II. Programme

What we will cover

Computer vision foundations

Core concepts and deep learning methods for analysing ecological imagery.

Ecological applications

Examples from camera traps, biodiversity monitoring and conservation.

iii

Hands-on practice

Practical exercises using real ecological datasets and modern tools.

From model to research

Discussing data, annotation, generalisation and responsible deployment.

The detailed programme will be announced soon.

III.Application

Who should apply?

We are looking for early-career researchers working in ecology, conservation, or related fields who are interested in using computer vision and machine learning to address real-world research or conservation challenges.

Master’s and PhD students, postdocs, early faculty, and junior researchers in ecology and conservation.
South African citizens or researchers currently based in South Africa with a valid visa allowing them to participate in the workshop.
Applicants with programming experience, preferably in Python. Python experience is not essential if you have some experience with R or another programming language and a clear plan to develop your Python skills before the workshop.
Applicants with a demonstrated interest in ecological or conservation research, or in deploying technology for conservation impact. A record of quantitative ecology research or technology deployment in conservation or government organisations is an advantage.
Applicants with a specific ecological or conservation question that computer vision or machine learning could help address.

Application materials

The workshop is limited to 20 participants. Applications will be evaluated through a selection process, with priority given to applicants who demonstrate a strong motivation to apply computer vision and machine learning to ecological or conservation challenges. Please prepare the following materials for your application.

1. Project proposal (maximum 500 words)

Describe an ecological or conservation problem you would like to address using computer vision or machine learning. Include:

Research problem: What question are you interested in, and how could computer vision or machine learning help?
Data availability: What data would you need? Do you already have access to them, or would you need to collect them? If you have data, describe their type, quantity and format (e.g. camera-trap images, drone footage or acoustic recordings). Please also indicate whether the data are already annotated or labelled, what types of annotations are available (e.g. species identification, bounding boxes, animal behaviours), and whether additional annotation would be needed. If so, briefly explain how you plan to carry it out. If you need to collect data, explain how and whether this is feasible.
Motivation and impact: What would success look like, and what could the impact be for science, conservation, policy or education?

2. Python readiness

Describe your programming experience. If you have Python experience, include a link to a GitHub repository containing Python code you have authored.

If you are not yet comfortable with Python, you may still apply. Please provide a concrete plan describing how you will develop your Python skills before the workshop.

3. Personal statement (maximum 1 page)

Describe your research experience, relevant skills, achievements and career goals, and explain how participating in the workshop would contribute to your work.

4. CV (maximum 2 pages)

Provide a concise CV highlighting your research experience, publications, technical skills, and relevant conservation or ecological work.

Optional: Letter of reference

A letter of reference is optional but may strengthen your application. Referees can send letters directly to cv4ecology-sa@aims.ac.za.

Go to application form

IV.Costs & support

There is no registration fee.

We are committed to making the workshop as accessible as possible and to ensuring that financial constraints are not a barrier to participation.

What is included?

Shared accommodation in Muizenberg for the duration of the workshop, arranged and covered by the organisers.
Meals at AIMS, including three meals per day and two tea/coffee breaks from Monday to Friday.
Arrival dinner and departure lunch.
Access to GPU-accelerated computing and storage throughout the workshop, with approximately 60 hours of dedicated compute time.
Five days of in-person lectures, practical sessions and group activities.
Small-group project work with dedicated instructors, organised around thematically similar ecological projects.
A collaborative learning environment for researchers working at the intersection of ecology, conservation and AI.

Travel and accommodation

Participants are expected to cover their own travel costs to and from AIMS. The organisers will arrange and cover shared accommodation for all selected participants.

If you prefer to stay in a single room or arrange your own accommodation, you are welcome to do so, but the booking and associated costs will be your responsibility.

Financial support for travel

We recognise that travel costs can be a significant barrier, particularly for students and early-career researchers. A limited amount of financial support will be available to help participants who are unable to cover their travel costs.

Requests will be considered on a case-by-case basis and subject to available funds, with priority given to applicants with the greatest financial need.

If you require travel support, indicate this in the main application form and complete the separate financial support application form.

Financial support form

V.Apply online

Ready to apply?

Use the main application form to submit your project proposal, Python readiness statement, personal statement and CV.

Main application form

If you need help with travel costs, indicate this in the main application form and submit the separate financial support form.

Financial support form

VI. Practical information

Plan your participation

Where: AIMS Muizenberg
Muizenberg, Cape Town, South Africa
When: 22–26 February 2027
Five-day workshop
Accommodation: Accommodation information will be provided to selected participants.
The organisers will arrange and cover shared accommodation. If you prefer a single room or arrange your own accommodation, the booking and costs will be your responsibility.
Funding: Limited travel support may be available.
Participants who require financial assistance will be able to apply for support. Awards will depend on available funding.

Logo