Metraipolis › archive › dc_b4919e69bf37
SPACE: Semantic Projection and Alignment of CLIP Embeddings for Domain Adaptation — arXiv:2609.23248
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.23248
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-09T21:59:29.701Z
- SHA-256 of the archived text
ee7d59e1f5bb067bfa0ffb623624dcb032dda890b661371429d00b68fbd505c8- Stable URL of this page
https://metraipolis.com/documento/dc_b4919e69bf37
How to cite
SPACE: Semantic Projection and Alignment of CLIP Embeddings for Domain Adaptation — arXiv:2609.23248 — arXiv (en), https://arxiv.org/abs/2609.23248 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b4919e69bf37 (Metraipolis archive dc_b4919e69bf37, sha256 ee7d59e1f5bb067bfa0ffb623624dcb032dda890b661371429d00b68fbd505c8)
BibTeX
@misc{dcb4919e69bf37,
title = {SPACE: Semantic Projection and Alignment of CLIP Embeddings for Domain Adaptation — arXiv:2609.23248},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.23248},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b4919e69bf37; sha256 ee7d59e1f5bb067bfa0ffb623624dcb032dda890b661371429d00b68fbd505c8},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.23248. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
A fundamental challenge in deploying vision models is domain shift, which arises when training and test data follow different distributions, leading to degraded performance. This challenge is amplified when the same semantic concept appears under distinct visual forms, such as photographs and sketches, where visual similarity is weak despite semantic correspondence. Existing unsupervised domain-adaptation methods aim to align distributions across domains but often ignore semantic relationships among samples of the same class. To address this issue, this paper introduces SPACE, a method that exploits the semantic structure of CLIP's vision-language space for domain adaptation. The key idea is to use text descriptions as semantic anchors by applying Singular Value Decomposition to CLIP embeddings of class descriptions, yielding an orthogonal basis that captures semantic relationships among categories. Visual features from both domains are projected into this semantic subspace, aligning images based on meaning rather than appearance. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-22 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.23248 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-10-09T21:45:20.867Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-22?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).