paper-with-me

홈 › Papers

USP: Unified Self-Supervised Pretraining for Image Generation and Understanding

2025-03-08 · Xiangxiang Chu, Renda Li, Yong Wang

Recent studies have highlighted the interplay between diffusion models and representation learning. Intermediate representations from diffusion models can be leveraged for downstream visual tasks, while self-supervised vision models can enhance the convergence and generation quality of diffusion models. However, transferring pretrained weights from vision models to diffusion models is challenging due to input mismatches and the use of latent spaces. To address these challenges, we propose Unified Self-supervised Pretraining (USP), a framework that initializes diffusion models via masked latent modeling in a Variational Autoencoder (VAE) latent space. USP achieves comparable performance in understanding tasks while significantly improving the convergence speed and generation quality of diffusion models. Our code will be publicly available at https://github.com/cxxgtxy/USP.

📄 PDF Abstract BibTeX arXiv:2503.06132

Code (1)

cxxgtxy/usp 공식 구현 pytorch

Tasks

Image GenerationRepresentation Learning

Methods 이 논문이 사용한 방법론

Diffusion Diffusion models generate samples by gradually removing noise from a signal, and their training objective can be expressed as a reweighted variational lower-bound…
SPEED The monocular depth estimation (MDE) is the task of estimating depth from a single frame. This information is an essential knowledge in many computer vision tasks such as scene…

Similar Papers 제목 키워드 기반

Unified Pretraining Framework for Document Understanding

2022-04-22 · Jiuxiang Gu, Jason Kuen, Vlad I. Morariu, Handong Zhao 외

Document intelligence automates the extraction of information from documents and supports many business applications. Recent self-supervised learning methods on large-scale unlabeled document datasets have opened up prom…

Document Layout Analysisdocument understandingSelf-Supervised Learning

UniDoc: Unified Pretraining Framework for Document Understanding

2021-12-01 · NeurIPS 2021 12 · Jiuxiang Gu, Jason Kuen, Vlad Morariu, Handong Zhao 외

Document intelligence automates the extraction of information from documents and supports many business applications. Recent self-supervised learning methods on large-scale unlabeled document datasets have opened up prom…

document understandingSelf-Supervised Learning

Self-Training Vision Language BERTs with a Unified Conditional Model

2022-01-06 · Xiaofeng Yang, Fengmao Lv, Fayao Liu, Guosheng Lin

Natural language BERTs are trained with language corpus in a self-supervised manner. Unlike natural language BERTs, vision language BERTs need paired data to train, which restricts the scale of VL-BERT pretraining. We pr…

Nucleus-aware Self-supervised Pretraining Using Unpaired Image-to-image Translation for Histopathology Images

2023-09-14 · Zhiyun Song, Penghui Du, Junpeng Yan, Kailu Li 외

Self-supervised pretraining attempts to enhance model performance by obtaining effective features from unlabeled data, and has demonstrated its effectiveness in the field of histopathology images. Despite its success, fe…

DiversityImage-to-Image TranslationInstance SegmentationMultiple Instance Learning+2

Learning Shared RGB-D Fields: Unified Self-supervised Pre-training for Label-efficient LiDAR-Camera 3D Perception

2024-05-28 · Xiaohao Xu, Ye Li, Tianyi Zhang, Jinrong Yang 외

Constructing large-scale labeled datasets for multi-modal perception model training in autonomous driving presents significant challenges. This has motivated the development of self-supervised pretraining strategies. How…

3D Object DetectionAutonomous DrivingNeRFNeural Rendering+4