Diffusion Model for Dense Matching

Fuente: arXiv
Enregistré dans:
Détails bibliographiques
Auteurs principaux: Nam, Jisu, Lee, Gyuseong, Kim, Sunwoo, Kim, Hyeonsu, Cho, Hyoungwon, Kim, Seyeon, Kim, Seungryong
Format: Preprint
Publié: 2023
Sujets:
Accès en ligne:
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866914651790376960
author Nam, Jisu
Lee, Gyuseong
Kim, Sunwoo
Kim, Hyeonsu
Cho, Hyoungwon
Kim, Seyeon
Kim, Seungryong
author_facet Nam, Jisu
Lee, Gyuseong
Kim, Sunwoo
Kim, Hyeonsu
Cho, Hyoungwon
Kim, Seyeon
Kim, Seungryong
contents The objective for establishing dense correspondence between paired images consists of two terms: a data term and a prior term. While conventional techniques focused on defining hand-designed prior terms, which are difficult to formulate, recent approaches have focused on learning the data term with deep neural networks without explicitly modeling the prior, assuming that the model itself has the capacity to learn an optimal prior from a large-scale dataset. The performance improvement was obvious, however, they often fail to address inherent ambiguities of matching, such as textureless regions, repetitive patterns, and large displacements. To address this, we propose DiffMatch, a novel conditional diffusion-based framework designed to explicitly model both the data and prior terms. Unlike previous approaches, this is accomplished by leveraging a conditional denoising diffusion model. DiffMatch consists of two main components: conditional denoising diffusion module and cost injection module. We stabilize the training process and reduce memory usage with a stage-wise training strategy. Furthermore, to boost performance, we introduce an inference technique that finds a better path to the accurate matching field. Our experimental results demonstrate significant performance improvements of our method over existing approaches, and the ablation studies validate our design choices along with the effectiveness of each component. Project page is available at https://ku-cvlab.github.io/DiffMatch/.
format Preprint
id arxiv_https___arxiv_org_abs_2305_19094
institution arXiv
publishDate 2023
record_format arxiv
spellingShingle Diffusion Model for Dense Matching
Nam, Jisu
Lee, Gyuseong
Kim, Sunwoo
Kim, Hyeonsu
Cho, Hyoungwon
Kim, Seyeon
Kim, Seungryong
Computer Vision and Pattern Recognition
The objective for establishing dense correspondence between paired images consists of two terms: a data term and a prior term. While conventional techniques focused on defining hand-designed prior terms, which are difficult to formulate, recent approaches have focused on learning the data term with deep neural networks without explicitly modeling the prior, assuming that the model itself has the capacity to learn an optimal prior from a large-scale dataset. The performance improvement was obvious, however, they often fail to address inherent ambiguities of matching, such as textureless regions, repetitive patterns, and large displacements. To address this, we propose DiffMatch, a novel conditional diffusion-based framework designed to explicitly model both the data and prior terms. Unlike previous approaches, this is accomplished by leveraging a conditional denoising diffusion model. DiffMatch consists of two main components: conditional denoising diffusion module and cost injection module. We stabilize the training process and reduce memory usage with a stage-wise training strategy. Furthermore, to boost performance, we introduce an inference technique that finds a better path to the accurate matching field. Our experimental results demonstrate significant performance improvements of our method over existing approaches, and the ablation studies validate our design choices along with the effectiveness of each component. Project page is available at https://ku-cvlab.github.io/DiffMatch/.
title Diffusion Model for Dense Matching
topic Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2305.19094