Results for ScanNet+FTSDF
Submission data
| Full name | ScanNet (re-implementation with flipped tsdf values) |
| Description | The method follows the volumetric segmentation approach described in:
ScanNet: Richly-annotated 3D Reconstructions of Indoor Scenes Angela Dai, Angel X. Chang, Manolis Savva, Maciej Halber, Thomas Funkhouser, Matthias Nießner However, we apply the following changes: 1. Flipped TSDF We use the flipped TSDF representation as described in: Semantic Scene Completion from a Single Depth Image Shuran Song Fisher Yu Andy Zeng Angel X. Chang Manolis Savva Thomas Funkhouser 2. We don't use free-space information. The only input to the network is a 61x31x31 volume with flipped TSDF values. 3. We only use the mesh as input The method computes a truncated distance field from the input mesh directly without using the provided depth images at all. |
| Publication authors | Martin Bokeloh, Juergen Sturm |
| Input Data Types | Uses Geometry Uses 3D |
| Programming language(s) | C++/Python/Tensorflow |
| Hardware | V100 |
| Submission creation date | 4 Sep, 2018 |
| Last edited | 5 Sep, 2018 |
3D semantic label results
| Info | avg iou | bathtub | bed | bookshelf | cabinet | chair | counter | curtain | desk | door | floor | otherfurniture | picture | refrigerator | shower curtain | sink | sofa | table | toilet | wall | window |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| 0.383 | 0.297 | 0.491 | 0.432 | 0.358 | 0.612 | 0.274 | 0.116 | 0.411 | 0.265 | 0.904 | 0.229 | 0.079 | 0.250 | 0.185 | 0.320 | 0.510 | 0.385 | 0.548 | 0.597 | 0.394 |