CDG-MAE: Learning Correspondences from Diffusion Generated Views

Fuente: arXiv
Salvato in:
Dettagli Bibliografici
Autori principali: Belagali, Varun, Marza, Pierre, Yellapragada, Srikar, Li, Zilinghan, Nandi, Tarak Nath, Madduri, Ravi K, Saltz, Joel, Christodoulidis, Stergios, Vakalopoulou, Maria, Samaras, Dimitris
Natura: Preprint
Pubblicazione: 2025
Soggetti:
Accesso online:
Tags: Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
_version_ 1866916806135906304
author Belagali, Varun
Marza, Pierre
Yellapragada, Srikar
Li, Zilinghan
Nandi, Tarak Nath
Madduri, Ravi K
Saltz, Joel
Christodoulidis, Stergios
Vakalopoulou, Maria
Samaras, Dimitris
author_facet Belagali, Varun
Marza, Pierre
Yellapragada, Srikar
Li, Zilinghan
Nandi, Tarak Nath
Madduri, Ravi K
Saltz, Joel
Christodoulidis, Stergios
Vakalopoulou, Maria
Samaras, Dimitris
contents Learning dense correspondences, critical for application such as video label propagation, is hindered by tedious and unscalable manual annotation. Self-supervised methods address this by using a cross-view pretext task, often modeled with a masked autoencoder, where a masked target view is reconstructed from an anchor view. However, acquiring effective training data remains a challenge - collecting diverse video datasets is difficult and costly, while simple image crops lack necessary pose variations. This paper introduces CDG-MAE, a novel MAE-based self-supervised method that uses diverse synthetic views generated from static images via an image-conditioned diffusion model. These generated views exhibit substantial changes in pose and perspective, providing a rich training signal that overcomes the limitations of video and crop-based anchors. We present a quantitative method to evaluate local and global consistency of generated images, discussing their use for cross-view self-supervised pretraining. Furthermore, we enhance the standard single-anchor MAE setting to a multi-anchor strategy to effectively modulate the difficulty of pretext task. CDG-MAE significantly outperforms state-of-the-art MAE methods reliant only on images and substantially narrows the performance gap to video-based approaches.
format Preprint
id arxiv_https___arxiv_org_abs_2506_18164
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle CDG-MAE: Learning Correspondences from Diffusion Generated Views
Belagali, Varun
Marza, Pierre
Yellapragada, Srikar
Li, Zilinghan
Nandi, Tarak Nath
Madduri, Ravi K
Saltz, Joel
Christodoulidis, Stergios
Vakalopoulou, Maria
Samaras, Dimitris
Computer Vision and Pattern Recognition
Learning dense correspondences, critical for application such as video label propagation, is hindered by tedious and unscalable manual annotation. Self-supervised methods address this by using a cross-view pretext task, often modeled with a masked autoencoder, where a masked target view is reconstructed from an anchor view. However, acquiring effective training data remains a challenge - collecting diverse video datasets is difficult and costly, while simple image crops lack necessary pose variations. This paper introduces CDG-MAE, a novel MAE-based self-supervised method that uses diverse synthetic views generated from static images via an image-conditioned diffusion model. These generated views exhibit substantial changes in pose and perspective, providing a rich training signal that overcomes the limitations of video and crop-based anchors. We present a quantitative method to evaluate local and global consistency of generated images, discussing their use for cross-view self-supervised pretraining. Furthermore, we enhance the standard single-anchor MAE setting to a multi-anchor strategy to effectively modulate the difficulty of pretext task. CDG-MAE significantly outperforms state-of-the-art MAE methods reliant only on images and substantially narrows the performance gap to video-based approaches.
title CDG-MAE: Learning Correspondences from Diffusion Generated Views
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2506.18164