Foundation Model or Finetune? Evaluation of few-shot semantic segmentation for river pollution
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866913492392476672 |
|---|---|
| author | Don, Marga Pinson, Stijn Cebrian, Blanca Guillen Asano, Yuki M. |
| author_facet | Don, Marga Pinson, Stijn Cebrian, Blanca Guillen Asano, Yuki M. |
| contents | Foundation models (FMs) are a popular topic of research in AI. Their ability to generalize to new tasks and datasets without retraining or needing an abundance of data makes them an appealing candidate for applications on specialist datasets. In this work, we compare the performance of FMs to finetuned pre-trained supervised models in the task of semantic segmentation on an entirely new dataset. We see that finetuned models consistently outperform the FMs tested, even in cases were data is scarce. We release the code and dataset for this work on GitHub. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2409_03754 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Foundation Model or Finetune? Evaluation of few-shot semantic segmentation for river pollution Don, Marga Pinson, Stijn Cebrian, Blanca Guillen Asano, Yuki M. Computer Vision and Pattern Recognition Foundation models (FMs) are a popular topic of research in AI. Their ability to generalize to new tasks and datasets without retraining or needing an abundance of data makes them an appealing candidate for applications on specialist datasets. In this work, we compare the performance of FMs to finetuned pre-trained supervised models in the task of semantic segmentation on an entirely new dataset. We see that finetuned models consistently outperform the FMs tested, even in cases were data is scarce. We release the code and dataset for this work on GitHub. |
| title | Foundation Model or Finetune? Evaluation of few-shot semantic segmentation for river pollution |
| topic | Computer Vision and Pattern Recognition |
| url | https://arxiv.org/abs/2409.03754 |