photo-import-setup

Photo Import & Project Setup

Overview

Every photogrammetry project starts with the same sequence: create project, import photos, load GPS, configure sensors, import masks, check image quality. This skill covers that entire setup phase, getting a project ready for alignment.

When to Use

• Starting a new project from raw photos
• Adding a new camera/capture to an existing project
• Re-importing after changing folder structure or GPS data
• Setting up sensors for the first time (fisheye, rolling shutter, etc.)

Prerequisites

• Photos on disk (EXR, JPEG, TIFF, PNG, DNG)
• GPS data (EXIF embedded or external CSV)
• Metashape MCP server running

Setup Pipeline

Step 1: Create or open project

create_project(path="E:\\Projects\\MyProject.psx")

Or for existing:

open_project(path="E:\\Projects\\MyProject.psx")

Step 2: Import photos

For a single folder:

add_photos(paths=["E:\\Captures\\Day1\\"])

For multiple folders (multi-day, multi-camera):

add_photos(paths=["E:\\Captures\\Day1\\", "E:\\Captures\\Day2\\"])

For glob patterns:

add_photos(paths=["E:\\Captures\\**\\*.exr"])

After import, verify:

list_chunks()

Check camera count matches expected photo count.

Step 3: Import GPS reference

If GPS is NOT in EXIF (common for vehicle-mounted captures):

import_reference(
    path="E:\\Captures\\gps_data.csv",
    columns="nxyz",        # n=label, x=longitude, y=latitude, z=altitude
    delimiter=","
)

Column format options:

• "nxyz" — label, lon, lat, alt (most common)
• "nxy" — label, lon, lat (no altitude)
• "nxyzabc" — label, lon, lat, alt, accuracy_x, accuracy_y, accuracy_z

After import, verify GPS loaded:

get_alignment_stats()

Check cameras_with_reference matches expected count.

Step 4: Configure sensors

This is the most critical setup step. Wrong sensor configuration is the #1 cause of total alignment failure.

Check current sensors:

list_sensors()