no code implementations • 16 Sep 2023 • Sungyeon Kim, Donghyun Kim, Suha Kwak
In this regard, we introduce a novel metric learning paradigm, called Universal Metric Learning (UML), which learns a unified distance metric capable of capturing relations across multiple data distributions.
no code implementations • ICCV 2023 • Junhyeong Cho, Gilhyun Nam, Sungyeon Kim, Hunmin Yang, Suha Kwak
In a joint vision-language space, a text feature (e. g., from "a photo of a dog") could effectively represent its relevant image features (e. g., from dog photos).
Ranked #1 on Domain Generalization on DomainNet
no code implementations • CVPR 2023 • Sungyeon Kim, Boseung Jeong, Suha Kwak
Supervision for metric learning has long been given in the form of equivalence between human-labeled classes.
1 code implementation • European Conference on Computer Vision (ECCV) 2022 • kyungmoon lee, Sungyeon Kim, Suha Kwak
Domain generalization is the task of learning models that generalize to unseen target domains.
Ranked #31 on Domain Generalization on Office-Home
no code implementations • 13 Aug 2022 • Sehyun Hwang, Sohyun Lee, Sungyeon Kim, Jungseul Ok, Suha Kwak
We consider the problem of active domain adaptation (ADA) to unlabeled target data, of which subset is actively selected and labeled given a budget constraint.
no code implementations • CVPR 2022 • Sungyeon Kim, Dongwon Kim, Minsu Cho, Suha Kwak
At the heart of our framework lies an algorithm that investigates contexts of data on the embedding space to predict their class-equivalence relations as pseudo labels.
no code implementations • 4 Jan 2022 • kyungmoon lee, Sungyeon Kim, Seunghoon Hong, Suha Kwak
Motivated by this, we introduce a new data augmentation approach that synthesizes novel classes and their embedding vectors.
no code implementations • 29 Sep 2021 • kyungmoon lee, Sungyeon Kim, Suha Kwak
For domain generalization, the task of learning a model that generalizes to unseen target domains utilizing multiple source domains, many approaches explicitly align the distribution of the domains.
Ranked #31 on Domain Generalization on Office-Home
2 code implementations • CVPR 2021 • Sungyeon Kim, Dongwon Kim, Minsu Cho, Suha Kwak
Our method exploits pairwise similarities between samples in the source embedding space as the knowledge, and transfers them through a loss used for learning target embedding models.
no code implementations • 1 Jan 2021 • Sungyeon Kim, Dongwon Kim, Minsu Cho, Suha Kwak
To this end, we design a new loss called smooth contrastive loss, which pulls together or pushes apart a pair of samples in a target embedding space with strength determined by their semantic similarity in the source embedding space; an analysis of the loss reveals that this property enables more important pairs to contribute more to learning the target embedding space.
3 code implementations • CVPR 2020 • Sungyeon Kim, Dongwon Kim, Minsu Cho, Suha Kwak
The former class can leverage fine-grained semantic relations between data points, but slows convergence in general due to its high training complexity.
Ranked #10 on Metric Learning on CUB-200-2011 (using extra training data)
Fine-Grained Image Classification Fine-Grained Vehicle Classification +1
1 code implementation • CVPR 2019 • Sungyeon Kim, Minkyo Seo, Ivan Laptev, Minsu Cho, Suha Kwak
Metric Learning for visual similarity has mostly adopted binary supervision indicating whether a pair of images are of the same class or not.