Kana Kim
< Back to selected work
2023 Data-Centric AI / Perception Evaluation

Autonomous Driving Dataset Quality Assurance

Automatically scoring and cleaning data quality without human review

PyTorchDetectron2 · YOLOv5/v7Ensemble votingLabeling · Whitening ScoreIoU · confidencePyQt5W&B
Autonomous Driving Dataset Quality Assurance Overview

Overview

After leading the collection and verification of a 930K-frame autonomous driving dataset, built ReliableDataset — a pipeline that removes noisy labels via model ensembles and whitens the data distribution to automatically score and prune the value of training data.

Autonomous Driving Dataset Quality Assurance Background

Background

Low-quality datasets yield inaccurate, biased, overfit, unfair, and unreliable models while wasting compute, time, and money — so dataset quality must be checked before training.

Autonomous Driving Dataset Quality Assurance Approach 1

Approach 1

Method I · Deleting noisy labels — models trained on split data cross-check each other’s labels (ensemble voting), and a Labeling Score based on IoU and confidence filters out wrong labels by threshold.

Autonomous Driving Dataset Quality Assurance Approach 2

Approach 2

Method II · Whitening the data distribution — a Diversity Score from class/box-size deviation (Z-score) is combined with the Labeling Score into a Whitening Score to sort and reduce data, finding the reduction ratio with the best accuracy.

Autonomous Driving Dataset Quality Assurance Result 1

Result 1

Before/after comparison — automatically corrected and cleaned wrong object labels for cars, pedestrians, and cyclists (e.g., ‘pedestrian’ → ‘cyclist’).

Autonomous Driving Dataset Quality Assurance Result 2

Result 2

Detection and label-cleaning results on collected data (day/night, buses, signs), compared before and after.