UperNet plus Swin backbone
The model card describes the UperNet semantic-segmentation framework leveraging a Swin Transformer backbone, with FPN and PPM components.
Open Source Model Profile · openmmlab
upernet-swin-base
upernet-swin-base is an openmmlab 121.93M image-segmentation model. Its card documents a UperNet plus Swin Transformer backbone.
- Publisher
- openmmlab
- Task
- image-segmentation
- Model type
- upernet
- License
- mit
- Library
- transformers
- Publication status
- Accepted · not indexed
Model overview
upernet-swin-base is published by openmmlab as an image-segmentation model. The captured configuration identifies UperNetForSemanticSegmentation with Safetensors metadata reporting 121,925,540 parameters. According to the model card, it combines the UperNet framework with a Swin Transformer base-sized backbone.
Recorded capabilities
121.93M segmentation weights
The captured configuration identifies UperNetForSemanticSegmentation and upernet, with Safetensors metadata reporting 121,925,540 parameters.
Documented paper lineage
According to the model card, the framework comes from the Unified Perceptual Parsing paper and the Swin combination from the Shifted Windows paper.
Use cases in the source record
- Semantic-segmentation experiments that use the documented UperNet backbone, FPN, and PPM structure.
- Vision-transformer comparisons that contrast the Swin-backed UperNet design with other segmentation backbones.
Limitations and unknowns
- No evaluation results were extracted from this record.
- No context-window value was extracted; none applies to this segmentation record beyond missing input-size documentation.
- Provider state is historical snapshot data and should be refreshed before being presented as current availability.
- The model card was written by the Hugging Face team rather than the releasing team, so publisher-level detail is limited.
Source and provenance
Source: openmmlab/upernet-swin-base
Captured: Unknown. Processed: 2026-09-07T19:34:54.356182+00:00.
UperNet, Swin Transformer base-sized backbone UperNet framework for semantic segmentation, leveraging a Swin Transformer backbone. UperNet was introduced in the paper Unified Perceptual Parsing for Scene Understanding by Xiao et al. Combining UperNet with a Swin Transformer backbone was introduced in the paper Swin Transformer: Hierarchical Vision Transformer using Shifted Windows . Disclaimer: The team releasing UperNet + Swin Transformer did not write a model card for this model so this model card has been written by the Hugging Face team. Model description UperNet is a framework for semantic segmentation. It consists of several c…
F001F002F003F004F005F006F007F008F009F010F011F012F013F014