Results for CSC-Pretrained
Submission data
Full name | Grouping Semantic Segmentation into Instances |
Description | Contrastive Scene Contexts Pre-trained SparseConv Backbone + Grouping based on semantic segmentation results (w/o using any additional modules, such as ScoreNet or Proposal Aggregation) |
Input Data Types | Uses Color,Uses Geometry Uses 3D |
Programming language(s) | PyTorch, CUDA, MinkovskiEngine |
Hardware | V100 |
Submission creation date | 21 Dec, 2018 |
Last edited | 22 Jan, 2021 |
3D semantic instance results
Info | avg ap | bathtub | bed | bookshelf | cabinet | chair | counter | curtain | desk | door | otherfurniture | picture | refrigerator | shower curtain | sink | sofa | table | toilet | window |
---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
0.405 | 0.738 | 0.465 | 0.331 | 0.205 | 0.655 | 0.051 | 0.601 | 0.092 | 0.211 | 0.329 | 0.198 | 0.459 | 0.775 | 0.195 | 0.524 | 0.400 | 0.878 | 0.184 |