首页/论文/2108.01775

arXiv 2108.017752022-02-04cs.CV

Solo-learn: A Library of Self-supervised Methods for Visual Representation Learning

University of Trento（特伦托大学）；SAP AI Research（SAP人工智能研究）；Fondazione Bruno Kessler（布鲁诺·凯斯勒基金会）

Victor G. Turrisi da Costa, Enrico Fini, Moin Nabi, Nicu Sebe, Elisa Ricci

备注：: Accepted to JMLR

阅读 PDF ↗

arXiv 页面 ↗HTML 原文 ↗

英文摘要相关论文

英文摘要：

This paper presents solo-learn, a library of self-supervised methods for visual representation learning. Implemented in Python, using Pytorch and Pytorch lightning, the library fits both research and industry needs by featuring distributed training pipelines with mixed-precision, faster data loading via Nvidia DALI, online linear evaluation for better prototyping, and many additional training tricks. Our goal is to provide an easy-to-use library comprising a large amount of Self-supervised Learning (SSL) methods, that can be easily extended and fine-tuned by the community. solo-learn opens up avenues for exploiting large-budget SSL solutions on inexpensive smaller infrastructures and seeks to democratize SSL by making it accessible to all. The source code is available at https://github.com/vturrisi/solo-learn.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

阅读原文

阅读 PDF ↗

arXiv 页面 ↗HTML 原文 ↗

本文内容

英文摘要相关论文

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

继续探索

相关论文

Smart Library: Identifying Books in a Library using Richly Supervised Deep Scene Text Reading

Pennsylvania State University（宾夕法尼亚州立大学）

Physical library collections are valuable and long standing resources for knowledge and learning. However, managing books in a large bookshelf and finding books on it often leads to tedious manual work, especially for large book collections where books might be missing or misplaced. Recently, deep neural models, such as Convolutional Neural Networks (CNN) and Recurrent Neural Networks (RNN) have achieved great success for scene text detection and recognition. Motivated by these recent successes, we aim to investigate their viability in facilitating book management, a task that introduces furth…

2016-11-22cs.CV

LIDARLearn: 一个用于3D点云分类、分割和自监督表示学习的统一深度学习库

LIDARLearn: A Unified Deep Learning Library for 3D Point Cloud Classification, Segmentation, and Self-Supervised Representation Learning

ENSIAS, Mohammed V University, Rabat, Morocco（穆罕默德五世大学ENSIAS学院，拉巴特，摩洛哥）

本文提出LIDARLearn，一个整合多种3D点云处理方法的统一深度学习库，支持分类、语义分割、部分分割和少样本学习，并提供标准化训练流程和严格多模型比较工具。

2026-04-12cs.CV

Semi-Supervised Endmember Identification In Nonlinear Spectral Mixtures Via Semantic Representation

University of Massachusetts（马萨诸塞大学）

This paper proposes a new hyperspectral unmixing method for nonlinearly mixed hyperspectral data using a semantic representation in a semi-supervised fashion, assuming the availability of a spectral reference library. Existing semi-supervised unmixing algorithms select members from an endmember library that are present at each of the pixels; most such methods assume a linear mixing model. However, those methods will fail in the presence of nonlinear mixing among the observed spectra. To address this issue, we develop an endmember selection method using a recently proposed semantic spectral rep…

2017-01-03cs.CV

SODAR: Segmenting Objects by DynamicallyAggregating Neighboring Mask Representations

National University of Singapore（新加坡国立大学）；Institute of Data Science, National University of Singapore（新加坡国立大学数据科学研究所）；NUS Graduate School of Integrative Sciences and Engineering（新加坡国立大学综合科学与工程研究生院）；YITU Technology（依图科技）

Recent state-of-the-art one-stage instance segmentation model SOLO divides the input image into a grid and directly predicts per grid cell object masks with fully-convolutional networks, yielding comparably good performance as traditional two-stage Mask R-CNN yet enjoying much simpler architecture and higher efficiency. We observe SOLO generates similar masks for an object at nearby grid cells, and these neighboring predictions can complement each other as some may better segment certain object part, most of which are however directly discarded by non-maximum-suppression. Motivated by the obse…

2022-12-23cs.CV

BlendTorch: A Real-Time, Adaptive Domain Randomization Library

Profactor GmbH（Profactor有限公司）；JKU（约翰内斯开普勒大学）

Solving complex computer vision tasks by deep learning techniques relies on large amounts of (supervised) image data, typically unavailable in industrial environments. The lack of training data starts to impede the successful transfer of state-of-the-art methods in computer vision to industrial applications. We introduce BlendTorch, an adaptive Domain Randomization (DR) library, to help creating infinite streams of synthetic training data. BlendTorch generates data by massively randomizing low-fidelity simulations and takes care of distributing artificial training data for model learning in re…

2020-10-06cs.CVcs.LG

MoViAD：一个用于视觉异常检测的模块化库

MoViAD: A Modular Library for Visual Anomaly Detection

University of Padova（帕多瓦大学）

MoViAD是一个模块化视觉异常检测库，提供先进模型、数据集与工具，支持多种学习场景及边缘部署，兼顾工程易用性与研究可扩展性。

2025-07-19cs.CV

更多相关论文收起更多论文

一种轻量级的用于基于能量的联合嵌入预测架构库

A Lightweight Library for Energy-Based Joint-Embedding Predictive Architectures

Meta FAIR；INRIA（法国国家信息与自动化研究所）；New York University（纽约大学）

本文提出EB-JEPA库，展示如何将图像级自监督学习的表示学习技术扩展到视频和动作条件世界模型，通过实验验证其在任务中的有效性。

2026-04-08cs.AIcs.CV

高光谱图像模型：技术报告

A PyTorch Library for Hyperspectral Image Models: Technical Report

Siddhartha Academy of Higher Education（悉达多高等教育学院）；Vellore Institute of Technology（韦洛尔理工学院）；Indira Gandhi National Open University（英迪拉·甘地国立开放大学）；Tezpur University（提斯浦尔大学）

提出高光谱图像模型框架，统一六种范式55个模型及24个基准场景，支持多种预处理与防重叠划分，通过1320次评估发现场景难度主导精度，无范式普遍占优。

2026-10-08cs.CV

XMP-Font: Self-Supervised Cross-Modality Pre-training for Few-Shot Font Generation

ByteDance Ltd（字节跳动有限公司）

Generating a new font library is a very labor-intensive and time-consuming job for glyph-rich scripts. Few-shot font generation is thus required, as it requires only a few glyph references without fine-tuning during test. Existing methods follow the style-content disentanglement paradigm and expect novel fonts to be produced by combining the style codes of the reference glyphs and the content representations of the source. However, these few-shot font generation methods either fail to capture content-independent style representations, or employ localized component-wise style representations, w…

2022-05-05cs.CV

射电天文学源分类的自监督学习：一项基准研究

Self-supervised learning for radio-astronomy source classification: a benchmark

University of Catania（卡塔尼亚大学）；INAF - Osservatorio Astrofisico di Catania（意大利国家天体物理研究所卡塔尼亚天文台）；Macquarie University（麦考瑞大学）；University of Milano Bicocca（米兰比可卡大学）

本文以SKA射电天文数据为背景，系统评估自监督学习在射电源分类中的效果，发现SSL在线性评估下显著优于传统预训练模型，并公开了模型与代码。

2024-11-22astro-ph.IMcs.CV

通过三模态一致性实现语言引导的视听源分离

Language-Guided Audio-Visual Source Separation via Trimodal Consistency

Boston University（波士顿大学）；Adobe Research（奥多比研究院）；MIT-IBM Watson AI Lab（麻省理工学院-IBM沃森人工智能实验室）；IBM Research（IBM研究院）

针对自然语言查询的视听源分离问题，提出仅用未标注音视频对、借助视觉-语言基础模型伪目标监督和三模态一致性损失的自监督方法，并在 MUSIC、SOLOS、AudioSet 上超过强监督方法。

2023-09-23cs.AIcs.CLcs.CV

S2Vec: 自监督的建成环境地理嵌入

S2Vec: Self-Supervised Geospatial Embeddings for the Built Environment

S2Vec通过自监督学习生成通用的地理空间嵌入，适用于建成环境特征的表示，并在社会经济任务中表现优异，同时支持多模态融合提升性能。

2026-01-07cs.AIcs.CVcs.SI

无共享门控或跨智能体梯度的自监督协作视觉专家群体中的涌现特化

Emergent Specialization in Populations of Self-Supervised Collaborative Vision Experts Without a Shared Gate or Cross-Agent Gradients

University of Bern（伯尔尼大学）

该研究证明，在没有共享门控或跨智能体梯度的条件下，自监督协作视觉专家群体可通过局部路由和无梯度通信涌现出有用的特化分工，提升性能并超越单一通才模型。

2026-09-30cs.AIcs.CVcs.LG

利用自监督学习在无标签、无注释的病理切片上绘制组织形态学癌症表型图谱

Mapping the landscape of histomorphological cancer phenotypes using self-supervised learning on unlabeled, unannotated pathology slides

University of Glasgow（格拉斯哥大学）；New York University School of Medicine（纽约大学医学院）；New York University Langone Health（纽约大学朗格尼医学中心）；Karolinska Institutet（卡罗林斯卡学院）；University College London Hospital（伦敦大学学院医院）

本文提出自监督方法HPL，无需注释即可从病理切片自动发现判别性特征，分组形成组织形态学表型库，揭示良性到恶性轨迹，并与肺癌患者生存及多癌种临床分子表型对齐。

2023-09-01cs.CVcs.LG

3D Self-Supervised Methods for Medical Imaging

Hasso-Plattner-Institute, Potsdam University Germany（哈索·普拉特纳研究所，波茨坦大学（德国））；Digital Health & Machine Learning（数字健康与机器学习（研究方向，非独立机构，不提取））

Self-supervised learning methods have witnessed a recent surge of interest after proving successful in multiple application fields. In this work, we leverage these techniques, and we propose 3D versions for five different self-supervised methods, in the form of proxy tasks. Our methods facilitate neural network feature learning from unlabeled 3D images, aiming to reduce the required cost for expert annotation. The developed algorithms are 3D Contrastive Predictive Coding, 3D Rotation prediction, 3D Jigsaw puzzles, Relative 3D patch location, and 3D Exemplar networks. Our experiments show that …

2020-11-02cs.CVcs.LGeess.IV

跨尺度MAE：遥感图像中的多尺度利用故事

Cross-Scale MAE: A Tale of Multi-Scale Exploitation in Remote Sensing

本文提出Cross-Scale MAE，通过多尺度增强和对比生成损失确保一致的表示，优于标准MAE及其他遥感MAE方法。

2024-01-29cs.AIcs.CVeess.IV