Citizen Science: first observations¶
At a glance
Register, log in, upload your first deployment, and review the AI-processed results in the Citizen Science frontend.
- Time: ~20 minutes plus upload time
- Who: new Citizen Science users — researchers, volunteers, NGO/park staff
- Prerequisites: an invitation to (or visibility of) a project on your TRAPPER instance
Getting started¶
1. Create an account¶
On the main login page, click Create an account.
- Fill in First name, Last name, E-mail, and a secure Password.
- Review and agree to the Terms of Service and Privacy Policy.
- Click Create an account to submit.
Important
Check your email for a verification link — you can't log in until you've activated your account.
Note
Already have a Trapper Expert account? Log in to the Citizen Science interface with the same email and password — no separate registration needed. If you still can't access a project, ask your project administrator for permissions.
2. Log in¶
Enter your email and password and click Sign in. Google/Microsoft sign-in is also available where an admin has configured it.
The main dashboard¶
After logging in you land on the project-selection page — every project you're a member of.
Click a project to open its dashboard: counts of locations, deployments, media files, recordings, and camera traps, plus a species-distribution bar chart.
Uploading camera trap media¶
1. Start a new upload¶
Main navigation → Upload.
2. Add deployment details¶
A deployment is the period a camera trap is installed at a location and actively recording, without being moved — the fundamental unit linking media to ecological analyses (occurrence, activity patterns, habitat use). Read more.
| Field | What it does | Example |
|---|---|---|
| Deployment name | Unique, descriptive identifier | bialowieza_lcm_2025 |
| Date range | Start/end of the deployment | Aug 1 – Sep 23, 2025 |
| Bait type | If any bait was used | none for most scientific studies |
3. Set the location¶
Click Create a new location, name it (e.g. bialowieza_loc1), then click the map to pin the exact spot — latitude/longitude fill in automatically.
Note
By default, the deployment name (code) combines with the location name into a unique identifier: {deployment_name}-{project_id}-{user_id}-{location_name}. Custom deployment IDs can be set via the Expert module — manually, or by importing a Camtrap-DP-format CSV (see Import locations & deployments). The only requirement is uniqueness within the project; deployments added this way become selectable in this upload form for users with the right permissions.
4. Camera and additional metadata (optional but recommended)¶
More data is always better for scientific analysis:
| Field | Example |
|---|---|
| Camera model | browning_btc6 |
| Camera ID | camera1 |
| Camera interval | 0 seconds between triggers |
| Camera height | 1.2 m |
| Camera tilt | -10 for a slight downward tilt; 0 = parallel to ground |
| Camera heading | 270 (compass degrees, West) |
| Detection distance | 20m estimated effective sensor range |
| Feature type | Road dirt, Water source, Carcass |
| Habitat | mixed forest, riparian zone |
| Timestamp issues | Check if the camera's clock was known to be wrong |
| Comments | Free text — e.g. "Camera knocked by a boar on Sept 15th" |
Camtrap DP compatible
All deployment metadata fields, from camera settings to location data, are fully compatible with the Camtrap DP standard — your data is structured, interoperable, and ready to share with global biodiversity platforms.
5. Upload your files¶
- Click the drag-and-drop / choose-file area.
- Select all the images/videos for this deployment (multiple files at once is fine).
- Wait for the progress bar to complete.
Automatic AI processing
Uploaded files automatically enter the AI pipeline (object detection, species classification) if one is configured on the project's Classification Project — see Create a classification project and AI pipeline architecture.
Viewing your data¶
Deployment summary¶
After upload and initial processing, the Deployment Summary page shows: a data panel (deployment ID, location, coordinates, date range, owner), a gallery (showing "Classification in progress…" while AI runs), an interactive map, extra metadata/comments and privacy/incomplete flags, and processing statistics.
The Images / Media view¶
Browse and filter all media files in the project — one row per file plus its observations.
Filter by Deployment, Species, Shared by team, Observation type, Sex, Age, and Classified by me.
The Image Viewer¶
Click any observation to open it: view AI bounding boxes/labels, zoom, adjust contrast/brightness/saturation (especially useful for nighttime infrared shots), navigate the sequence via thumbnails, and play video inline.
Classifying your first sequence¶
Classification view (in the left sidebar) is where you review, correct, and validate the AI's work.
Sequences¶
Media is grouped into sequences — an ecological event, e.g. an animal or group passing the camera. The system groups media captured within a configurable time window (default 5 minutes) into one sequence, so you analyze one event instead of many separate photos. Read more.
Navigate at three levels: single object (click a bounding box, or hotkeys), single media (next/previous, thumbnails), or whole event/sequence (sequence strip, jump to ID).
Keyboard shortcuts
Press Ctrl + / (Cmd + / on macOS) for the full shortcut list. See Video annotation & interpolation for the complete reference if you're annotating video.
The interactive canvas¶
The image viewer is a workspace: draw new bounding boxes by clicking and dragging, select/resize/move existing ones, zoom and pan. On-the-fly Contrast/Brightness/Saturation sliders help in challenging lighting — dark infrared shots or overexposed daytime images.
Annotation modes¶
| Mode | What it does |
|---|---|
| Single Object | Click one bounding box to classify just that object — most precise |
| All Objects on Image | Classify everything in the current view at once |
| Group Classification (bulk) | Select multiple thumbnails in the sequence strip; one classification applies to every object in every selected image — fastest for sequences of the same species |
The classification form¶
The form is dynamic and project-specific — fields (species, sex, age, count, behaviour) are not hard-coded. A project manager defines them via the Classificator. observation_type and species are always required; others (age, sex, custom fields) are optional. All predefined fields conform to Camtrap DP.
Custom fields can carry a data type (string/integer/float/boolean) and validation rules — e.g. a wolf project might add howling_activity (boolean) and health_status (string | enum); a roe deer project might add antler_development (boolean). Custom fields map onto the Camtrap DP schema (e.g. as observationTags) so datasets stay shareable across projects.
Managing teams & your profile¶
Creating a team¶
Teams are a self-organization tool for coordinating fieldwork and annotation across researchers, volunteers, NGO/park staff — local collaboration while project permissions stay governed by administrators.
- Go to Teams, create a new one.
- Name and describe it.
- Draw a polygon (or upload GeoJSON/GPX) for the team's area of interest.
Profile and settings¶
Click your profile icon to manage: General (email, name, phone, institution, bio), Change password, Multi-factor authentication, Delete account, and Language (English, Swedish, Polish, German).
Next steps¶
- Video annotation & interpolation — for video sequences
- Bounding boxes for ML training data
- Create a classification project — for project managers configuring the Classificator and AI pipeline