Few Shot Semantic Segmentation On Coco 20I 5
Metrics
FB-IoU
Mean IoU
Results
Performance results of various models on this benchmark
Comparison Table
| Model Name | FB-IoU | Mean IoU |
|---|---|---|
| mianet-aggregating-unbiased-instance-and-1 | 73.81 | 51.03 |
| eliminating-feature-ambiguity-for-few-shot | 78.5 | 57.1 |
| hmfs-hybrid-masking-for-few-shot-segmentation | 72.2 | 49.4 |
| doubly-deformable-aggregation-of-covariance | 71.6 | 48.1 |
| apanet-adaptive-prototypes-alignment-network | - | 43 |
| self-calibrated-cross-attention-network-for | 74.2 | 53.9 |
| iterative-few-shot-semantic-segmentation-from | - | 44.4 |
| unleashing-the-potential-of-the-diffusion | - | 60.7 |
| visual-and-textual-prior-guided-mask-assemble | 76.7 | 57.1 |
| dense-gaussian-processes-for-few-shot | - | 56.2 |
| few-shot-segmentation-without-meta-learning-a | - | 41.6 |
| prototype-as-query-for-few-shot-semantic | 73.3 | 53.4 |
| feature-weighting-and-boosting-for-few-shot | - | 23.65 |
| fecanet-boosting-few-shot-semantic | 71.1 | 47.6 |
| cost-aggregation-with-4d-convolutional-swin | 72.4 | 47.9 |
| apanet-adaptive-prototypes-alignment-network | - | 43.2 |
| clustered-patch-element-connection-for-few | - | 46.9 |
| self-support-few-shot-semantic-segmentation | - | 50.2 |
| simpler-is-better-few-shot-semantic | - | 41.3 |
| masked-cross-image-encoding-for-few-shot | - | 51.04 |
| self-support-few-shot-semantic-segmentation | - | 44.1 |
| mianet-aggregating-unbiased-instance-and-1 | 73.13 | 51.65 |
| prototype-as-query-for-few-shot-semantic | 74.6 | 54.7 |
| few-shot-semantic-segmentation-with-support | 66.2 | 48 |
| interclass-prototype-relation-for-few-shot | - | 51.1 |
| feature-proxy-transformer-for-few-shot | - | 58.9 |
| doubly-deformable-aggregation-of-covariance | 72.9 | 49.2 |
| dense-gaussian-processes-for-few-shot | - | 57.9 |
| msdnet-multi-scale-decoder-for-few-shot | 74.5 | 54.5 |
| part-aware-prototype-network-for-few-shot | - | 38.5 |
| cost-aggregation-is-all-you-need-for-few-shot | - | 47.9 |
| hypercorrelation-squeeze-for-few-shot | 72.4 | 49.5 |
| panet-few-shot-image-semantic-segmentation | 63.5 | 29.7 |
| hypercorrelation-squeeze-for-few-shot | 70.7 | 46.9 |
| mining-latent-classes-for-few-shot | - | 41.4 |
| hmfs-hybrid-masking-for-few-shot-segmentation | 72.2 | 48.4 |
| apanet-adaptive-prototypes-alignment-network | - | 46.4 |
| hmfs-hybrid-masking-for-few-shot-segmentation | 73.3 | 50.6 |
| visual-and-textual-prior-guided-mask-assemble | 79.4 | 61.8 |
| bridge-the-points-graph-based-few-shot | - | 66.8 |
| hmfs-hybrid-masking-for-few-shot-segmentation | 71.8 | 48.3 |
| singular-value-fine-tuning-few-shot-1 | - | 54.38 |
| msanet-multi-similarity-and-attention-1 | 56.8 | 56.3 |
| beyond-the-prototype-divide-and-conquer | - | 46.48 |
| dense-cross-query-and-support-attention | 71.7 | 48.3 |
| hybrid-mamba-for-few-shot-segmentation | 77.6 | 58.9 |
| hmfs-hybrid-masking-for-few-shot-segmentation | 72.9 | 50.6 |
| msanet-multi-similarity-and-attention-1 | 53.67 | 50.47 |
| self-calibrated-cross-attention-network-for | 74.8 | 57 |
| anti-aliasing-semantic-reconstruction-for-few | - | 35.75 |
| hybrid-mamba-for-few-shot-segmentation | 75.5 | 54.5 |
| singular-value-fine-tuning-few-shot-1 | - | 53.87 |
| eliminating-feature-ambiguity-for-few-shot | 74.3 | 52.8 |
| clustered-patch-element-connection-for-few | - | 46.7 |
| msdnet-multi-scale-decoder-for-few-shot | 75.1 | 55.3 |
| mining-latent-classes-for-few-shot | - | 45.1 |
| dense-cross-query-and-support-attention | 76.9 | 58.3 |
| seggpt-segmenting-everything-in-context | - | 67.9 |
| self-guided-and-cross-guided-learning-for-few | - | 39.9 |
| learning-what-not-to-segment-a-new | - | 51.16 |
| hierarchical-dense-correlation-distillation | - | 52.4 |
| prior-guided-feature-enrichment-network-for | 61.9 | 37.4 |
| quaternion-valued-correlation-learning-for | - | 50 |
| feature-weighting-and-boosting-for-few-shot | - | 22.63 |
| simpler-is-better-few-shot-semantic | - | 42 |
| prior-guided-feature-enrichment-network-for | 61.6 | 37.7 |
| dense-cross-query-and-support-attention | 73.3 | 51.9 |
| hierarchical-dense-correlation-distillation | 77.7 | 56 |
| integrative-few-shot-learning-for | - | 49.5 |
| feature-proxy-transformer-for-few-shot | - | 53.8 |
| quaternion-valued-correlation-learning-for | - | 51.9 |
| matcher-segment-anything-with-one-shot-using | - | 60.7 |
| fecanet-boosting-few-shot-semantic | 67.7 | 41.5 |
| singular-value-fine-tuning-few-shot-1 | - | 49.49 |
| singular-value-fine-tuning-few-shot-1 | - | 49.07 |
| intermediate-prototype-mining-transformer-for | - | 47.5 |
| adaptive-prototype-learning-and-allocation | 66.96 | 42.48 |
| few-shot-segmentation-via-cycle-consistent | - | 45.6 |
| intermediate-prototype-mining-transformer-for | - | 47.9 |
| prototype-mixture-models-for-few-shot | - | 35.5 |
| interclass-prototype-relation-for-few-shot | - | 53.3 |