Ir para o conteúdo
ModelListadosExterno
Baixar

Ccip Anime Character Id

CCIP enables determining whether two images depict the same character, calculating visual similarity between characters.

Fontehuggingface.co/nebulette/ccip-anime-character-id/tree/2c45301b74fe07a86178a064042d31f41ffed3e6
TipoOtherVersãoModelCnota

Fonte do modelo

Tipo
Other
Tamanho do arquivo
36,4 MB
Versão
vModel
Atualizado
10 de out. de 2026
Fonte
Hugging Face

Descrição da fonte

Character similarity between images

CCIP enables determining whether two images depict the same character, calculating visual similarity between characters.

There are up to 6000 (female) characters available.

The model assigns a score to each image patch to make the features distinguishable on a 2D plane.

The cosine similarity between the scores of two images is used to identify the same characters in both.

The takeaway is that the model doesn't need a tokenizer, text model, or cross-attention. The labels are unknown to the model during training.

Source data

  • live gelbooru feed
  • character_index
  • character_similarity

References

  • Yann LeCun's $1B Bet Against LLMs [Welch Labs]

Fontes

1 fonte
Hugging FacePrimárioVersão 2c45301b74fe07a86178a064042d31f41ffed3e6
huggingface.co/nebulette/ccip-anime-character-id/tree/2c45301b74fe07a86178a064042d31f41ffed3e6

Verificado 10 de out.

Artefatos de modelo

1 artefato

model.safetensors

safetensors · 36,4 MB · SHA-256 4b9a3efa3df4…c88e · Hugging Face

Download file

Trechos de fonte

2 trechos
Readme1 de ago.huggingface.co/nebulette/ccip-anime-character-id

--- license: apache-2.0 tags: - anime-tagger - character pipeline_tag: image-feature-extraction --- Character similarity between images CCIP enables determining whether two images depict the same character, calculating visual similarity between characters. There are up to 6000 (female) characters available. The model assigns a score to each image patch to make the features distinguishable on a 2D plane. ![](images/bell_shape.png) The cosine similarity between the scores of two images is used to identify the same characters in both. The takeaway is that the model doesn't need a tokenizer, text model, or cross-attention. The labels are unknown to the model during training. Source data - live gelbooru feed - character_index - character_similarity References - Yann LeCun's $1B Bet Against LLMs [Welch Labs]

Note1 de ago.huggingface.co/nebulette/ccip-anime-character-id

Source context: 0 downloads · 0 likes · Pipeline image-feature-extraction · Repo nebulette/ccip-anime-character-id