Semantic Segmentation on PASTIS
Mean IoU (test)
- 2021-07-16 — U-TAE: Mean IoU (test) 63.1
- 2023-05-03 — Exchanger+Mask2Former: Mean IoU (test) 67.9
| Rank | Model | Mean IoU (test) | Number of Params | Overall Accuracy | Paper | Code | Year |
|---|---|---|---|---|---|---|---|
| 1 | Exchanger+Mask2Former | 67.9 | – | – | Revisiting the Encoding of Satellite Image Time Series | TotalVariation/Exchanger4SITS | 2023 |
| 2 | Exchanger+Unet | 66.8 | – | – | Revisiting the Encoding of Satellite Image Time Series | TotalVariation/Exchanger4SITS | 2023 |
| 3 | U-TAE | 63.1 | 1.1M | 83.2 | Panoptic Segmentation of Satellite Image Time Series with Convolutional Temporal Attention Networks | VSainteuf/utae-paps | 2021 |
| 4 | Exchanger+Mask2Former | 67.9 | – | – | Revisiting the Encoding of Satellite Image Time Series | TotalVariation/Exchanger4SITS | 2023 |
| 5 | Exchanger+Unet | 66.8 | – | – | Revisiting the Encoding of Satellite Image Time Series | TotalVariation/Exchanger4SITS | 2023 |
| 6 | U-TAE | 63.1 | 1.1M | 83.2 | Panoptic Segmentation of Satellite Image Time Series with Convolutional Temporal Attention Networks | VSainteuf/utae-paps | 2021 |