Semantic Segmentation Annotation Case Study

Nerd For Tech
Published in
2 min readApr 6, 2023


What is semantic segmentation?

Semantic segmentation is a computer vision task that involves assigning a semantic label to every pixel in an image. It is a form of dense prediction, where the goal is to classify each and every pixel in the input image, as opposed to object detection, where the goal is to identify and classify individual objects in the image. The semantic labels can correspond to object categories, such as “person” or “car”, or to scene elements, such as “sky” or “road”.

There are several types of semantic segmentation, including:

Instance Segmentation: This type of semantic segmentation involves separating different objects of the same category, such as separating different people or cars in an image.

Semantic Segmentation for Medical Imaging: In medical imaging, semantic segmentation is used to segment different organs, tissues, and other structures in medical images such as CT scans or MRIs.

Panoptic Segmentation: This is a combination of instance and semantic segmentation, where the goal is to identify and segment both objects and stuff (e.g. background) in an image.

Scene Parsing: This type of semantic segmentation involves assigning labels to different elements of a scene, such as sky, road, buildings, and vegetation.

Salient Object Segmentation: This type of semantic segmentation focuses on identifying and segmenting the most noticeable or important objects in an image.

Overall, the type of semantic segmentation depends on the specific task and the type of data being analyzed.

Now let’s have a look at a semantic segmentation case.


Segmentation of all water in the picture (flooding, surface puddles)

Segmentation Criteria

  • Segmentation of all standing water
  • Determine whether it is standing water by successive frames
  • For fences (fences, nets, etc.) that contain many small holes, if the water is visible in the naked eye, the visible part needs to be segmented.

Data Output

xml, json


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?



Nerd For Tech

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