文本挖掘和预处理工具（文本清洗、新词发现、情感分析、实体识别链接、关键词抽取、知识抽取、句法分析等），无监督或弱监督方法
(CVPR'20 Oral) Unsupervised Learning of Probably Symmetric Deformable 3D Objects from Images in the Wild
Pytorch version of SfmLearner from Tinghui Zhou et al.
Machine learning movie recommending system
Official PyTorch implementation of SynDiff described in the paper (https://arxiv.org/abs/2207.08208).
A Non-Autoregressive Transformer based Text-to-Speech, supporting a family of SOTA transformers with supervised and unsupervised duration modelings. This project grows with the research community, aiming to achieve the ultimate TTS
CVPR2018: Unsupervised Cross-dataset Person Re-identification by Transfer Learning of Spatio-temporal Patterns
This project proposes an end-to-end framework for semi-supervised Anomaly Detection and Segmentation in images based on Deep Learning.
Unsupervised Deep Homography: A Fast and Robust Homography Estimation Model
Official PyTorch code for WACV 2022 paper "CFLOW-AD: Real-Time Unsupervised Anomaly Detection with Localization via Conditional Normalizing Flows"
DualGAN-tensorflow: tensorflow implementation of DualGAN
A high performance impermentation of Unsupervised Image Segmentation by Backpropagation - Asako Kanezaki
Dynamically get the suggested clusters in the data for unsupervised learning.
Official implementation of SuperSimpleNet [ICPR 2024, JIMS 2025]
(ECCV 2022) Code for Share With Thy Neighbors: Single-View Reconstruction by Cross-Instance Consistency
A Non-Autoregressive End-to-End Text-to-Speech (text-to-wav), supporting a family of SOTA unsupervised duration modelings. This project grows with the research community, aiming to achieve the ultimate E2E-TTS
[ICCV 2021] Excavating the Potential Capacity of Self-Supervised Monocular Depth Estimation
This repository tries to provide unsupervised deep learning models with Pytorch
本项目实现了一种基于 VAE-CycleGAN 的图像重建无监督缺陷检测算法。该算法结合了变分自编码器 (VAE) 和 CycleGAN 的优势，无需标注数据即可检测图像中的缺陷/异常。This project implements an unsupervised defect detection algorithm for image reconstruction based on VAE-CycleGAN. This algorithm combines the advantages of variational autoencoders (VAE) and CycleGAN to detect defects in images without any supervision.
Joint Disentangling and Adaptation for Cross-Domain Person Re-Identification. ECCV'20 (Oral)
To associate your repository with the unsupervised topic, visit your repo's landing page and select "manage topics."