Skip to content
ModelListedExternal
Download

Segformer B1 Nlver Finetuned 1024 1024 Cir Preview

This is a segformer segmentation model finetuned on random samples from the entire 41,000 square kilometer of aerial photography data, see pdok aerial data, and using the BGT, see pdok BGT data. Specifically it takes...

Sourcehuggingface.co/jgerbscheid/segformer_b1-nlver_finetuned-1024-1024-cir-preview/tree/c9f5add58a90263b941d53e78800d9913cdb130b
KindOtherBase modelnvidia/segformer-b1-finetuned-cityscapes-1024-1024VersionRevision c9f5add

Model source

Kind
Other
File size
52.2 MB
Base model
nvidia/segformer-b1-finetuned-cityscapes-1024-1024
Version
vRevision c9f5add
Updated
Sep 27, 2026
Downloads
13
Tasks
Image segmentation
Source
Hugging Face

Source description

NL-aerial segmentation

This is a segformer segmentation model finetuned on random samples from the entire 41,000 square kilometer of aerial photography data, see pdok aerial data, and using the BGT, see pdok BGT data. Specifically it takes in 1024x1024 CIR aerial photographs taken at a resolution of 8cm/pixel and predicts water, buildings, roads/pavement, vegetation. This model is part of the NL-veranderdetectie project, summary in dutch. The model was trained using this codebase, see the repo for more details.

Model Details

Regular segformer, with the classification head adjusted to 5 classes.

Model Description

  • Developed by: Het Waterschapshuis, in collaboration with the Dutch Waterboards [More Information Needed]
  • Model type: [Segformer Semantic Segmentation Model]
  • License: [MIT]
  • Finetuned from model: [nvidia/segformer-b1-finetuned-cityscapes-1024-1024]

Model Sources [optional]

Uses

This model is used to create a segmentation map of the dutch waterways based on aerial imagery. This segmentation map is then used to actualize the BGT, a national map maintained by the Dutch government.

Training Hyperparameters

See the nlveranderdetectie repostiory for details of the training setup. https://gitlab.com/hetwaterschapshuis/kenniscentrum/tooling/nlveranderdetectie/ , specifically look at the yaml configs in the run/configs directory.

Hardware

The model was trained on a single nvidia 3090 GPU for ~2 days or ~126k forward passes with a batch size of 8.

Model Card Authors

The model car was written by J. Gerbscheid, email: j.gerbscheid@hetwaterschapshuis.nl

Model Card Contact

The model was trained by J. Gerbscheid, email: j.gerbscheid@hetwaterschapshuis.nl

Sources

1 source
Hugging FacePrimaryVersion c9f5add58a90263b941d53e78800d9913cdb130b
huggingface.co/jgerbscheid/segformer_b1-nlver_finetuned-1024-1024-cir-preview/tree/c9f5add58a90263b941d53e78800d9913cdb130b

Verified Sep 27

Model artifacts

1 artifact

model.safetensors

safetensors · 52.2 MB · SHA-256 61155e0eb5c1…a90a · Hugging Face

Download file

Source excerpts

2 excerpts
ReadmeSep 27huggingface.co/jgerbscheid/segformer_b1-nlver_finetuned-1024-1024-cir-preview

--- library_name: transformers tags: - aerial_photography - segmentation - netherlands - cir license: mit base_model: - nvidia/segformer-b1-finetuned-cityscapes-1024-1024 pipeline_tag: image-segmentation --- # NL-aerial segmentation This is a segformer segmentation model finetuned on random samples from the entire 41,000 square kilometer of aerial photography data, see [pdok aerial data](https://www.pdok.nl/introductie/-/article/pdok-luchtfoto-rgb-open-), and using the BGT, see [pdok BGT data](https://www.pdok.nl/introductie/-/article/basisregistratie-grootschalige-topografie-bgt-). Specifically it takes in 1024x1024 CIR aerial photographs taken at a resolution of 8cm/pixel and predicts water, buildings, roads/pavement, vegetation. This model is part of the NL-veranderdetectie project, [summary](https://www.winnovatie.nl/kennisbank/3041030.aspx?t=Learning-paper-NL-Veranderdetectie-Fase-1) in dutch. The model was trained using this [codebase](https://gitlab.com/hetwaterschapshuis/kenniscentrum/tooling/nlveranderdetectie/), see the repo for more details. ## Model Details Regular segformer, with the classification head adjusted to 5 classes. ### Model Description - **Developed by**: Het [Waterschapshuis](https://www.hetwaterschapshuis.nl/), in collaboration with the Dutch Waterboards [More Information Needed] <!-- - **Funded by [optional]:** [] --> <!-- - **Shared by [optional]:** [More Information Needed] --> - **Model type:** [Segformer Semantic Segmentation Model] - **License:** [MIT] - **Finetuned from model:** [nvidia/segformer-b1-finetuned-cityscapes-1024-1024] ### Model Sources [optional] <!-- Provide the basic links for the model. --> - **Repository:** [https://gitlab.com/hetwaterschapshuis/kenniscentrum/tooling/nlveranderdetectie/](https://gitlab.com/hetwaterschapshuis/kenniscentrum/tooling/nlveranderdetectie/) <!-- - **Paper [optional]:** [More Information Needed] - **Demo [optional]:** [More Information Needed] --> ## Uses This model is used to create a segmentation map of the dutch waterways based on aerial imagery. This segmentation map is then used to actualize the BGT, a national map maintained by the Dutch government. #### Training Hyperparameters See the nlveranderdetectie repostiory for details of the training setup. [https://gitlab.com/hetwaterschapshuis/kenniscentrum/tooling/nlveranderdetectie/](https://gitlab.com/hetwaterschap...

NoteSep 27huggingface.co/jgerbscheid/segformer_b1-nlver_finetuned-1024-1024-cir-preview

Source context: 13 downloads · 0 likes · Pipeline image-segmentation · Library transformers · Repo jgerbscheid/segformer_b1-nlver_finetuned-1024-1024-cir-preview