ProtoSeg: Interpretable Semantic Segmentation with Prototypical Parts
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2023
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866909662608097280 |
|---|---|
| author | Sacha, Mikołaj Rymarczyk, Dawid Struski, Łukasz Tabor, Jacek Zieliński, Bartosz |
| author_facet | Sacha, Mikołaj Rymarczyk, Dawid Struski, Łukasz Tabor, Jacek Zieliński, Bartosz |
| contents | We introduce ProtoSeg, a novel model for interpretable semantic image segmentation, which constructs its predictions using similar patches from the training set. To achieve accuracy comparable to baseline methods, we adapt the mechanism of prototypical parts and introduce a diversity loss function that increases the variety of prototypes within each class. We show that ProtoSeg discovers semantic concepts, in contrast to standard segmentation models. Experiments conducted on Pascal VOC and Cityscapes datasets confirm the precision and transparency of the presented method. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2301_12276 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | ProtoSeg: Interpretable Semantic Segmentation with Prototypical Parts Sacha, Mikołaj Rymarczyk, Dawid Struski, Łukasz Tabor, Jacek Zieliński, Bartosz Computer Vision and Pattern Recognition We introduce ProtoSeg, a novel model for interpretable semantic image segmentation, which constructs its predictions using similar patches from the training set. To achieve accuracy comparable to baseline methods, we adapt the mechanism of prototypical parts and introduce a diversity loss function that increases the variety of prototypes within each class. We show that ProtoSeg discovers semantic concepts, in contrast to standard segmentation models. Experiments conducted on Pascal VOC and Cityscapes datasets confirm the precision and transparency of the presented method. |
| title | ProtoSeg: Interpretable Semantic Segmentation with Prototypical Parts |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2301.12276 |