Fishyscapes is a public benchmark for uncertainty estimation in a real-world task of semantic segmentation for urban driving. It evaluates pixel-wise uncertainty estimates towards the detection of anomalous objects in front of the vehicle.
Variants: Fishyscapes, Fishyscapes L&F
This dataset is used in 1 benchmark:
| Task | Model | Paper | Date |
|---|---|---|---|
| Anomaly Detection | Mask2Anomaly | Unmasking Anomalies in Road-Scene Segmentation | 2023-07-25 |
| Anomaly Detection | FlowEneDet | Concurrent Misclassification and Out-of-Distribution Detection … | 2023-05-16 |
| Anomaly Detection | RPL+CoroCL | Residual Pattern Learning for Pixel-wise … | 2022-11-26 |
| Anomaly Detection | DenseHybrid | DenseHybrid: Hybrid Anomaly Detection for … | 2022-07-06 |
| Anomaly Detection | PEBAL | Pixel-wise Energy-biased Abstention Learning for … | 2021-11-24 |
| Anomaly Detection | SML | Standardized Max Logits: A Simple … | 2021-07-23 |
| Anomaly Detection | Synboost | Pixel-wise Anomaly Detection in Complex … | 2021-03-09 |
| Anomaly Detection | Bayesian DeepLab | Evaluating Bayesian Deep Learning Methods … | 2018-11-30 |
Recent papers with results on this dataset: