Nerd For Tech
Published in

Nerd For Tech

Autonomous driving Annotation Case Study — Semantic Segmentation in 3D Point Cloud

ByteBridge 3D Semantic Segmentation Annotation Project

3D Semantic Segmentation

3D Semantic segmentation is to classify each pixel in the image, determine each category, and then divide areas.

In autonomous driving, segmentation annotation can help vehicles understand the surrounding road environment well, such as roads, vehicles, pedestrians, green plants, and other obstacles.

Next, let’s have a look at a 3D semantic segmentation annotation project.

Project requirements:

  1. Labeling instruction

(1) Unlabeled data

If the original/unlabeled data is poor in quality or the original image is blurred, please do not label it.

If the images show up repeatedly (such as parking situations, except traffic jam), only label the first frame.

(2) Annotation category

The cuboid should match the outline of the subject within 2 pixels.

Blurred objects with pixels less than 10×10 pixels don’t need to be labeled separately, but they can be labeled with the tag “other obstacles”.

If an object is less than 10x10 pixels but its boundary is clearly visible, it should be labeled.

If the category of an object cannot be determined, it can be labeled with the tag “other obstacles.

No need for imagination, only label visible obstacles and their visible part (except transparent part, if the occlusion part takes up more than 80%, there is no need to label the object).

2. Labelling Categories


Outsource your data labeling tasks to ByteBridge, you can get high-quality ML training datasets cheaper and faster!

  • Free Trial Without Credit Card: you can get your sample result in a fast turnaround, check the output, and give feedback directly to our project manager.
  • 100% Human Validated
  • Transparent & Standard Pricing: clear pricing is available(labor cost included)

Why not have a try?



Get the Medium app

A button that says 'Download on the App Store', and if clicked it will lead you to the iOS App store
A button that says 'Get it on, Google Play', and if clicked it will lead you to the Google Play store



Data labeling outsourced service: get your ML training datasets cheaper and faster!—