Matterport3D: Learning from RGB-D data in indoor environments

Keyphrases

• Indoor Environment 100%
• RGB-D 100%
• Existing Data 20%
• Semantic Segmentation 20%
• Scene Understanding 20%
• Surface Reconstruction 20%
• Building Scale 20%
• Keypoint Matching 20%
• Computer Vision Tasks 20%
• Camera Pose 20%
• 3D Semantic Segmentation 20%
• Number of Views 20%
• Real-time Semantic Segmentation 20%
• Global Alignment 20%
• Semantic Regions 20%
• Region Classification 20%
• Panoramic View 20%
• Color Semantics 20%
• Segmentation Classification 20%

Computer Science

• Image Segmentation 100%
• Annotation 50%
• Scene Understanding 50%
• Surface Reconstruction 50%
• Computer Vision Task 50%
• RGB-D Image 50%