Metraipolis › archive › dc_a096e2b004f2
PixelUp: Zero-Shot Semantic Feature Upsampling for Fine-Grained Vision Tasks — arXiv:2608.02792
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2608.02792
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T12:58:04.269Z
- SHA-256 of the archived text
8270b8867f670a9ec9d7e2b50b2d4b24cf9e7df59e0053fa6c7ba6b4d4760b90- Stable URL of this page
https://metraipolis.com/documento/dc_a096e2b004f2
How to cite
PixelUp: Zero-Shot Semantic Feature Upsampling for Fine-Grained Vision Tasks — arXiv:2608.02792 — arXiv (en), https://arxiv.org/abs/2608.02792 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_a096e2b004f2 (Metraipolis archive dc_a096e2b004f2, sha256 8270b8867f670a9ec9d7e2b50b2d4b24cf9e7df59e0053fa6c7ba6b4d4760b90)
BibTeX
@misc{dca096e2b004f2,
title = {PixelUp: Zero-Shot Semantic Feature Upsampling for Fine-Grained Vision Tasks — arXiv:2608.02792},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2608.02792},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_a096e2b004f2; sha256 8270b8867f670a9ec9d7e2b50b2d4b24cf9e7df59e0053fa6c7ba6b4d4760b90},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.02792. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Self-supervised Vision Foundation Models (VFMs) have become essential backbones for downstream tasks due to their strong and transferable visual representations. However, their patch-token-level features are often too coarse for dense prediction tasks such as semantic segmentation and depth estimation when accurate fine-grained predictions are required. Feature upsampling methods have been developed to recover pixel-level detail but still face limitations. Learnable upsamplers are often designed for a specific encoders and must be retrained for different encoders. Image-guided methods that use shallow pixel encoders often introduce textural artifacts and lack the semantic guidance needed for accurate downstream predictions. We introduce PixelUp, a zero-shot VFM-agnostic upsampler achieving semantic awareness through a coarse-to-fine chain of windowed cross-attention architecture guided by multi-scale semantic features. We demonstrate that PixelUp outperforms both VFM-specific and VFM-agnostic upsamplers, achieving state-of-the-art performance on dense prediction tasks with an average improvement of +1.2 mIoU on semantic segmentation and +0.25 $\delta_1$, on NYUv2 depth estimation across VFMs. PixelUp further improves training-free open-vocabulary and unsupervised semantic segmentation by an average of +1.3 mIoU and +0.5 mIoU, respectively. Code available at this https URL Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-08-05 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2608.02792 — 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 2026-10-31T12:45:33.198Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-08-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).