Metraipolis › archive › dc_c1cc427ee202
FAVE: Foveated Adaptive Visual Encoding for Efficient Fine-Grained Visual Understanding — arXiv:2609.04392
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2609.04392
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-16T23:33:20.012Z
- SHA-256 of the archived text
3522f35de2e9c0da3fda8a82b29aec1379876a92858ee05e30e805a0e00ec674- Stable URL of this page
https://metraipolis.com/documento/dc_c1cc427ee202
How to cite
FAVE: Foveated Adaptive Visual Encoding for Efficient Fine-Grained Visual Understanding — arXiv:2609.04392 — arXiv (en), https://arxiv.org/abs/2609.04392 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_c1cc427ee202 (Metraipolis archive dc_c1cc427ee202, sha256 3522f35de2e9c0da3fda8a82b29aec1379876a92858ee05e30e805a0e00ec674)
BibTeX
@misc{dcc1cc427ee202,
title = {FAVE: Foveated Adaptive Visual Encoding for Efficient Fine-Grained Visual Understanding — arXiv:2609.04392},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2609.04392},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_c1cc427ee202; sha256 3522f35de2e9c0da3fda8a82b29aec1379876a92858ee05e30e805a0e00ec674},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2609.04392. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Fine-grained visual understanding depends on local detail, yet visual encoders face a trade-off between costly full-image high-resolution processing and compact global encoding that can weaken such evidence. Inspired by human active vision, we separate where to look from what to encode. We focus on the latter and introduce FAVE (Foveated Adaptive Visual Encoding), a lightweight variable-resolution ViT that encodes externally selected regions at high acuity while preserving native geometry. We first isolate this encoding problem using oracle ground-truth crops in a controlled small-object regime. On ImageNet objects with a native maximum side of 96 pixels, FAVE improves Top-1 by 9.4 points over a fixed-resolution ViT on the same crop window with 12.7 times lower FLOPs. Increasing global resolution or backbone capacity does not recover the same operating point. We then integrate FAVE as a complementary local branch in FastVLM. Its local tokens are combined with FastVLM's global visual tokens, while the original global pathway and language model remain frozen. With at most 16 additional local tokens, FAVE improves TextVQA by 1.60 points and achieves a 3.3 times controlled TTFT speedup over SmolVLM2-2.2B. On GQA attribute questions, it improves FastVLM-1.5B by 1.31 points, extending the benefit beyond text while narrowing the gap to FastVLM-7B. Together, these results show that selectively allocating high-acuity local capacity provides an efficient complement to broader global representations and model scaling for fine-grained understanding of small objects, text, and attributes. Tema degli esploratori: statistica e scienza dei dati (retro-catalogo arXiv: annunci del 2026-09-07 in cs.CV). Categorie arXiv: Computer Vision and Pattern Recognition (cs.CV). Pagina del lavoro: https://arxiv.org/abs/2609.04392 — 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 2026-10-16T23:16:11.381Z dalla pagina https://arxiv.org/catchup/cs.CV/2026-09-07?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).