There are 160 categories of indoor scenes in Places and 67 categories in MIT67 in total.
For Places, it contains both indoor and outdoor scenes. We extracted the indoor parts from the original dataset based on the label provided by the authors. More details about Places can be found here. And for MIT67, original data and labels can be found in this website.
The two scene-centric datasets have 57 categories in common.
In this repository, files with the same number in their names indicate overlapping categories contained by both MIT67 and Places. Such as names_mit_10.txt and names_places_10.txt contain same 10 categories of indoor scenes. The same for names_mit_35.txt and names_places_35.txt, etc.
Other details and data will be updated in the future.
- https://register.apple.com/resources/imdf/Categories/#venue
- Venue Types
- Airport
- Shopping Center
- Train Station
- 2020 | Indoor mapping and modeling by parsing floor plan images | [paper]
- 2019 | ICCV | Deep floor plan recognition using a multi-task network with room-boundary-guided attention | [paper] [project]
- 2019 | CubiCasa5K: A Dataset and an Improved Multi-Task Model for Floorplan Image Analysis | [paper] [project]
- CubiCasa5K is a large-scale floorplan image dataset containing 5000 samples annotated into over 80 floorplan object categories. The dataset annotations are performed in a dense and versatile manner by using polygons for separating the different objects.
- 2017 | ICCV | Raster-to-Vector: Revisiting Floorplan Transformation | [paper] [project]
- 2017 | Parsing floor plan images | [paper]
- 2013 | Statistical segmentation and structural recognition for floor plan interpretation | [paper]
- CVC-FP: Database for structural floor plan analysis
- 2021 | ICCV | 3D-FRONT: 3D Furnished Rooms With layOuts and semaNTics | [paper project]
- 2021 | CVPR | Plan2Scene: Converting floorplans to 3D scenes | [paper] [code]
- 2019 | ICCV | Floor-SP: Inverse CAD for Floorplans by Sequential Room-wise Shortest Path | [paper] [project] [code]
- 2019 | ICCV | Floorplan-Jigsaw: Jointly Estimating Scene Layout and Aligning Partial Scans | [paper]
- 2016 | CVPR | 3D Semantic Parsing of Large-Scale Indoor Spaces | [paper]
- 2015 | ICCV | Structured Indoor Modeling | [paper] [project]
- 2015 | CVPR | Rent3D: Floor-Plan Priors for Monocular Layout Estimation | [paper] [project]