# Overview

There's a lot of useful information that comes out of a successful calibration:

• A new plex.json that contains the optimized Plex of the system
• A metrics.json file with the optimized object space of the dataset, as well as the calibration metrics and statistics for the overall calibration process, including every component

Both of these files can be found in and downloaded from the Calibration Detail page produced after every successful calibration.

!!! info "Calibration Metrics Schema" Get an idea of what this Metrics JSON file contains and how it is structured at the Calibration Metrics Schema documentation.

## Optimized Plex

The optimized Plex is a description of your now-calibrated System. This Plex is typically more "complete" and information-rich than the input Plex, since it is based off of the real data used to calibrate by the System.

!!! info "The Plex" Read more about the Plex structure and utility in the docs.

## Optimized Object Space

TVCal will optimize the object space of every calibration. For example, If your object space consists of a checkerboard, TVCal will directly measure how flat (or not) the checkerboard actually is using the calibration data. This effect can be visualized and used to improve calibrations going forward.

!!! warning "Coming Soon" This page is under construction! Check back soon. In the meantime, one can find optimized object space data for every calibration in its metrics file.

## Calibration Metrics and Statistics

This is where the fun starts. Every time TVCal processes a calibration, it produces a ton of useful data about that calibration's overall performance and quality.

### Summary Statistics

For every calibration process, there are a few critical outputs (for instance, variance) that can provide insight into calibration quality. However, these outputs can be complex to understand, and they should be interpreted holistically with the rest of the data provided. Deep dive into these outputs and impacts here.

### Camera-Specific Metrics

Every image input into TVCal that detects a sufficient portion of the object space will be used in the optimization. For each of these images, there will be an associated set of image measurements, image pose data (world extrinsic + covariance), and corresponding reprojection errors (residuals). This page describes some of the common ways TVCal handles and visualizes these metrics.