{ "cells": [ { "cell_type": "markdown", "id": "d0d2e007", "metadata": {}, "source": [ "# Find Label Errors in Semantic Segmentation Datasets\n", "\n", "This 5-minute quickstart tutorial shows how you can use cleanlab to find potentially mislabeled images in semantic segmentation datasets. In semantic segmentation, our data consists of images each annotated with a corresponding mask that labels each pixel in the image as one of K classes. Models are trained on this labeled mask to predict the class of each pixel in an image. However in real-world data, this annotated mask often contains errors. \n", "Here we apply cleanlab to find label errors in a variant of the [SYNTHIA](https://synthia-dataset.net) segmentation dataset, which consists of synthetic images generated via graphics engine." ] }, { "cell_type": "markdown", "id": "07936a54", "metadata": {}, "source": [ "