162 citations · 364 across the 26 of their papers we have counts for
19 papers · 1 filter
VisDA 2022 Challenge: Domain Adaptation for Industrial Waste Sorting
Dina Bashkirova, Samarth Mishra, Diala Lteif +16
Label-efficient and reliable semantic segmentation is essential for many real-life applications, especially for industrial settings with high visual diversity, such as waste sortin…
Going Beyond Nouns With Vision & Language Models Using Synthetic Data
Paola Cascante-Bonilla, Khaled Shehada, James Seale Smith +8
Large-scale pre-trained Vision & Language (VL) models have shown remarkable performance in many applications, enabling replacing a fixed set of supported classes with zero-shot ope…
Exploring Consistency in Cross-Domain Transformer for Domain Adaptive Semantic Segmentation
Kaihong Wang, Donghyun Kim, Rogerio Feris +2
While transformers have greatly boosted performance in semantic segmentation, domain adaptive transformers are not yet well explored. We identify that the domain gap can cause disc…
CODA-Prompt: COntinual Decomposed Attention-based Prompting for Rehearsal-Free Continual Learning
James Seale Smith, Leonid Karlinsky, Vyshnavi Gutta +6
Computer vision models suffer from a phenomenon known as catastrophic forgetting when learning novel concepts from continuously shifting training data. Typical solutions for this c…
Teaching Structured Vision&Language Concepts to Vision&Language Models
Sivan Doveh, Assaf Arbelle, Sivan Harary +8
Vision and Language (VL) models have demonstrated remarkable zero-shot performance in a variety of tasks. However, some aspects of complex language understanding still remain a cha…
Grafting Vision Transformers
Jongwoo Park, Kumara Kahatapitiya, Donghyun Kim +3
Vision Transformers (ViTs) have recently become the state-of-the-art across many computer vision tasks. In contrast to convolutional networks (CNNs), ViTs enable global information…
Temporal Relevance Analysis for Video Action Models
Quanfu Fan, Donghyun Kim, Chun-Fu +4
In this paper, we provide a deep analysis of temporal modeling for action recognition, an important but underexplored problem in the literature. We first propose a new approach to…
A Unified Framework for Domain Adaptive Pose Estimation
Donghyun Kim, Kaihong Wang, Kate Saenko +2
While pose estimation is an important computer vision task, it requires expensive annotation and suffers from domain shift. In this paper, we investigate the problem of domain adap…
A Broad Study of Pre-training for Domain Generalization and Adaptation
Donghyun Kim, Kaihong Wang, Stan Sclaroff +1
Deep models must learn robust and transferable representations in order to perform well on new domains. While domain transfer methods (e.g., domain adaptation, domain generalizatio…
Learning Cross-modal Contrastive Features for Video Domain Adaptation
Donghyun Kim, Yi-Hsuan Tsai, Bingbing Zhuang +4
Learning transferable and domain adaptive feature representations from videos is important for video-relevant tasks such as action recognition. Existing video domain adaptation met…
Tune it the Right Way: Unsupervised Validation of Domain Adaptation via Soft Neighborhood Density
Kuniaki Saito, Donghyun Kim, Piotr Teterwak +3
Unsupervised domain adaptation (UDA) methods can dramatically improve generalization on unlabeled target domains. However, optimal hyper-parameter selection is critical to achievin…
OpenMatch: Open-set Consistency Regularization for Semi-supervised Learning with Outliers
Kuniaki Saito, Donghyun Kim, Kate Saenko
Semi-supervised learning (SSL) is an effective means to leverage unlabeled data to improve a model's performance. Typical SSL methods like FixMatch assume that labeled and unlabele…
Self-supervised Visual Attribute Learning for Fashion Compatibility
Donghyun Kim, Kuniaki Saito, Samarth Mishra +3
Many self-supervised learning (SSL) methods have been successful in learning semantically meaningful visual representations by solving pretext tasks. However, prior work in SSL foc…
Cross-domain Self-supervised Learning for Domain Adaptation with Few Source Labels
Donghyun Kim, Kuniaki Saito, Tae-Hyun Oh +3
Existing unsupervised domain adaptation methods aim to transfer knowledge from a label-rich source domain to an unlabeled target domain. However, obtaining labels for some source d…
Universal Domain Adaptation through Self Supervision
Kuniaki Saito, Donghyun Kim, Stan Sclaroff +1
Unsupervised domain adaptation methods traditionally assume that all source categories are present in the target domain. In practice, little may be known about the category overlap…
MILA: Multi-Task Learning from Videos via Efficient Inter-Frame Attention
Donghyun Kim, Tian Lan, Chuhang Zou +5
Prior work in multi-task learning has mainly focused on predictions on a single image. In this work, we present a new approach for multi-task learning from videos via efficient int…
MULE: Multimodal Universal Language Embedding
Donghyun Kim, Kuniaki Saito, Kate Saenko +2
Existing vision-language methods typically support two languages at a time at most. In this paper, we present a modular approach which can easily be incorporated into existing visi…
Multi-way Encoding for Robustness
Donghyun Kim, Sarah Adel Bargal, Jianming Zhang +1
Deep models are state-of-the-art for many computer vision tasks including image classification and object detection. However, it has been shown that deep models are vulnerable to a…
Semi-supervised Domain Adaptation via Minimax Entropy
Kuniaki Saito, Donghyun Kim, Stan Sclaroff +2
Contemporary domain adaptation methods are very effective at aligning feature distributions of source and target domains without any target supervision. However, we show that these…