跳到正文
研踪PaperTrail
← 研踪 PaperTrail

arXiv:2611.24806 · v1

ReCoCo: From Patches to Compositions for Visual Representation Learning

ReCoCo:从图像块到组合的视觉表征学习

首次提交 2026-11-19 01:45
版本更新 2026-11-19 01:45
雷达观测 2026-11-19 15:04
arXiv ↗PDF ↗

摘要

Self-supervised learning has produced powerful visual representations, yet learning strong dense features remains challenging. While consistency between augmented views provides strong image-level supervision, extending this principle to local visual content is less straightforward. Existing approaches often match individual patches across views or identify coherent regions, introducing uncertainties in correspondence estimation and assumptions about how visual content should be grouped. We introduce ReCoCo (Regional Composition Consistency), a principled approach that extends cross-view consistency to the regional level without requiring such correspondences or semantic grouping. ReCoCo represents the content shared between augmented views through compositions of their patch-level predictions over learned visual prototypes. By aligning these compositions, the model learns from regions containing multiple visual concepts without requiring them to be separated beforehand. ReCoCo integrates naturally with teacher-student self-distillation and introduces no additional learned parameters. Extensive experiments demonstrate substantial improvements in frozen-feature segmentation, clustering, and regional retrieval while maintaining competitive image-level recognition, alleviating the trade-off between dense and global representation quality.

AI 论文解读

基于摘要 · 非全文审阅

ReCoCo 通过在学习到的视觉原型上对齐增强视图间图像块预测的组合,把跨视图一致性扩展到区域级,无需块级对应或语义分组。

正在读取账号与研究记录…

版本差异

目前只收录了一个版本。收录下一版后,可以对照标题、作者、分类和摘要。

观测到的版本

相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。

v1 · 2026-11-19 01:45 · 最新观测版本

ReCoCo: From Patches to Compositions for Visual Representation Learning

Self-supervised learning has produced powerful visual representations, yet learning strong dense features remains challenging. While consistency between augmented views provides strong image-level supervision, extending this principle to local visual content is less straightforward. Existing approaches often match individual patches across views or identify coherent regions, introducing uncertainties in correspondence estimation and assumptions about how visual content should be grouped. We introduce ReCoCo (Regional Composition Consistency), a principled approach that extends cross-view consistency to the regional level without requiring such correspondences or semantic grouping. ReCoCo represents the content shared between augmented views through compositions of their patch-level predictions over learned visual prototypes. By aligning these compositions, the model learns from regions containing multiple visual concepts without requiring them to be separated beforehand. ReCoCo integrates naturally with teacher-student self-distillation and introduces no additional learned parameters. Extensive experiments demonstrate substantial improvements in frozen-feature segmentation, clustering, and regional retrieval while maintaining competitive image-level recognition, alleviating the trade-off between dense and global representation quality.

查看此版本解读 →查看此版本原文 ↗