Figure - PMC

Skip to main content

An official website of the United States government

Here's how you know

Here's how you know

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you've safely connected to the .gov website. Share sensitive information only on official, secure websites.

View full-text article in PMC

. 2024 Nov 20;6:176–182. doi: 10.1109/OJEMB.2024.3503499

Search in PMC
Search in PubMed
View in NLM Catalog
Add to search

© 2024 The Authors

This work is licensed under a Creative Commons Attribution 4.0 License. For more information, see https://creativecommons.org/licenses/by/4.0/

PMC Copyright notice

Fig. 5. — Improved predictions on the held-out test data. From left to right: input images ( column), ground truths ( column), SAM masks ( column), predictions from ( column), and final binary occlusion segmentation mask ( column). In the and columns, white regions are classified as occlusions while black ones are classified as non-occlusions. In the column, different colors simply indicate different image segments without associated occlusion class, which is the main shortcoming of SAM in our task. Green boxes highlight occlusion easily visible segmentation improvements.

Inline graphic — Improved predictions on the held-out test data. From left to right: input images ( column), ground truths ( column), SAM masks ( column), predictions from ( column), and final binary occlusion segmentation mask ( column). In the and columns, white regions are classified as occlusions while black ones are classified as non-occlusions. In the column, different colors simply indicate different image segments without associated occlusion class, which is the main shortcoming of SAM in our task. Green boxes highlight occlusion easily visible segmentation improvements.