Skip to content
finwave Docs

Quick Start

This page walks you from “I have nothing” to “I have a submitted encounter that the ML pipeline is processing.” It should take about 10 minutes.

What you’ll learn

Section titled “What you’ll learn”
  • How to create an account and sign in
  • How to be added to (or join) a population
  • How to submit your first encounter
  • What happens after submission

1. Create an account

Section titled “1. Create an account”

Go to app.finwave.io and sign up with your email address. You’ll receive a verification email — click the link to activate the account, then sign in.

Your account is independent of any population. Once it exists, you can be invited into one or more populations and one or more organisations.

2. Join a population

Section titled “2. Join a population”

You need to be a member of at least one population before you can do useful work. The fastest path is the /populations index page — every population on finwave is listed there, each with a state badge that tells you exactly how to get in:

  • 🔵 Open to join — click Join population. You’re in immediately, no admin approval needed.
  • 🟠 Request to join — click Request to join, optionally add a short message. An administrator reviews and approves.
  • 🟢 Member — you’re already in. Click Open population to switch your workspace into it.
  • ⚪ Request pending — you’ve already submitted a request; wait for the admin to respond.

If you were sent an email invitation, the population appears in your workspace switcher (top-left of the sidebar) automatically — you don’t need to visit /populations first.

If you’re not sure which population to join, ask the person who pointed you at finwave. Most working groups have a designated administrator.

3. Submit your first encounter

Section titled “3. Submit your first encounter”

From the sidebar, choose Encounters → New Encounter. Three things are required:

  1. Date — when the observation happened.
  2. Location — drop a pin on the map, or enter coordinates and a name.
  3. At least one image — drag and drop your photos into the upload area.

Everything else (organisation, predation flag, behaviours, license) is optional but improves the research value of your data. See Creating Encounters for a full walkthrough of every field.

Click Submit. You’ll see a per-image upload progress card. When all images have been received, the encounter card shows the Upload complete status.

4. Watch the ML pipeline work

Section titled “4. Watch the ML pipeline work”

After submission, finwave automatically pushes your images through three ML stages:

  1. Detection — the model finds candidate animals in each photo (typically by dorsal fin).
  2. Classification — body side (left / right / unknown) is assigned to each detection.
  3. Identification — each detection is compared against the population’s catalogue of known individuals.

You don’t have to do anything to trigger this. Status indicators on the encounter overview update as each phase completes — typically a few minutes for a small encounter.

5. Review the results

Section titled “5. Review the results”

Once the pipeline has finished, open the encounter and click Annotator. You’ll see:

  • Bounding boxes the model drew.
  • The model’s top guesses for each box, with confidence.
  • A path to confirm correct identifications, correct wrong ones, or add ones the model missed.

Confirmed identifications are what turn raw photos into long-term population data. See Annotator Overview for how the review interface works.

What’s next?

Section titled “What’s next?”