What question did this study set out to answer?

This research aims to enhance the segmentation of indoor scenes by utilizing depth information alongside RGB inputs.

synapse

⌘+K

synapse

⌘+K

January 16, 2013Open Access

Indoor Semantic Segmentation using depth information

Key Points

This research aims to enhance the segmentation of indoor scenes by utilizing depth information alongside RGB inputs.
Implemented a multiscale convolutional network to learn features from RGB-D images.
Evaluated performance using the NYU-v2 depth dataset.
Demonstrated real-time labeling of indoor scenes in video sequences.
Achieved state-of-the-art accuracy of 64.5% on the NYU-v2 depth dataset.
Enabled processing of indoor scene segmentation in real-time with appropriate hardware.

Abstract

This work addresses multi-class segmentation of indoor scenes with RGB-D inputs. While this area of research has gained much attention recently, most works still rely on hand-crafted features. In contrast, we apply a multiscale convolutional network to learn features directly from the images and the depth information. We obtain state-of-the-art on the NYU-v2 depth dataset with an accuracy of 64.5%. We illustrate the labeling of indoor scenes in videos sequences that could be processed in real-time using appropriate hardware such as an FPGA.

Read Full Paperexternally

Mark Helpful

Bookmark

Relay

View Full Paper