A high-resolution semantic segmentation dataset with 50 validation and 100 test objects. Image resolution in BIG ranges from 2048×1600 to 5000×3600. Every image in the dataset has been carefully labeled by a professional while keeping the same guidelines as PASCAL VOC 2012 without the void region.
Variants: BIG
This dataset is used in 1 benchmark:
| Task | Model | Paper | Date |
|---|---|---|---|
| Semantic Segmentation | PSPNet + CascadePSP | CascadePSP: Toward Class-Agnostic and Very … | 2020-05-06 |
| Semantic Segmentation | RefineNet + CascadePSP | CascadePSP: Toward Class-Agnostic and Very … | 2020-05-06 |
| Semantic Segmentation | DeepLabV3+ + CascadePSP | CascadePSP: Toward Class-Agnostic and Very … | 2020-05-06 |
| Semantic Segmentation | FCN + CascadePSP | CascadePSP: Toward Class-Agnostic and Very … | 2020-05-06 |
Recent papers with results on this dataset: