SegNeXt-B

Reported on 6 benchmarks across 2 tasks · 1 paper

Note: results are matched by exact model name. Different papers may use the same name for different model variants.

Medical3 results

Semantic SegmentationonDSEC
mIoU· 2022-09-18
71.55
best: 74.94 (BRENet)
SegNeXt: Rethinking Convolutional Attention Design for Semantic Segmentation arXiv:2209.08575
Semantic SegmentationonDDD17
mIoU· 2022-09-18
71.46
best: 78.56 (BRENet)
SegNeXt: Rethinking Convolutional Attention Design for Semantic Segmentation arXiv:2209.08575
Semantic SegmentationoniSAID
mIoU· 2022-09-18
69.9
best: 70.3 (SegNeXt-L)
SegNeXt: Rethinking Convolutional Attention Design for Semantic Segmentation arXiv:2209.08575

Audio3 results

10-shot image generationonDSEC
mIoU· 2022-09-18
71.55
best: 74.94 (BRENet)
SegNeXt: Rethinking Convolutional Attention Design for Semantic Segmentation arXiv:2209.08575
10-shot image generationonDDD17
mIoU· 2022-09-18
71.46
best: 78.56 (BRENet)
SegNeXt: Rethinking Convolutional Attention Design for Semantic Segmentation arXiv:2209.08575
10-shot image generationoniSAID
mIoU· 2022-09-18
69.9
best: 70.3 (SegNeXt-L)
SegNeXt: Rethinking Convolutional Attention Design for Semantic Segmentation arXiv:2209.08575