Hallucination improves the performance of unsupervised visual representation learning

J Wu, J Hobbs, N Hovakimyan - Proceedings of the IEEE …, 2023 - openaccess.thecvf.com
Contrastive learning models based on Siamese structure have demonstrated remarkable
performance in self-supervised learning. Such a success of contrastive learning relies on …

Margin-based few-shot class-incremental learning with class-level overfitting mitigation

Y Zou, S Zhang, Y Li, R Li - Advances in neural information …, 2022 - proceedings.neurips.cc
Few-shot class-incremental learning (FSCIL) is designed to incrementally recognize novel
classes with only few training samples after the (pre-) training on base classes with sufficient …

SGMNet: Scene graph matching network for few-shot remote sensing scene classification

B Zhang, S Feng, X Li, Y Ye, R Ye… - IEEE Transactions on …, 2022 - ieeexplore.ieee.org
Few-shot remote sensing scene classification (FSRSSC) is an important task, which aims to
recognize novel scene classes with few examples. Recently, several studies attempt to …

Beyond rgb: Scene-property synthesis with neural radiance fields

M Zhang, S Zheng, Z Bao, M Hebert… - Proceedings of the …, 2023 - openaccess.thecvf.com
Comprehensive 3D scene understanding, both geometrically and semantically, is important
for real-world applications such as robot perception. Most of the existing work has focused …

Task encoding with distribution calibration for few-shot learning

J Zhang, X Zhang, Z Wang - … on Circuits and Systems for Video …, 2022 - ieeexplore.ieee.org
Few-shot learning is an extremely challenging task in computer vision that has attracted
increased research attention in recent years. However, most recent methods do not fully use …

Multi-task View Synthesis with Neural Radiance Fields

S Zheng, Z Bao, M Hebert… - Proceedings of the IEEE …, 2023 - openaccess.thecvf.com
Multi-task visual learning is a critical aspect of computer vision. Current research, however,
predominantly concentrates on the multi-task dense prediction setting, which overlooks the …

TDNet: A novel transductive learning framework with conditional metric embedding for few-shot remote sensing image scene classification

B Wang, Z Wang, X Sun, Q He… - IEEE Journal of …, 2023 - ieeexplore.ieee.org
Few-shot learning, which aims to learn the concept of novel category from extremely limited
labeled samples, has received intense interests in remote sensing image scene …

Generative modeling for multi-task visual learning

Z Bao, M Hebert, YX Wang - International Conference on …, 2022 - proceedings.mlr.press
Generative modeling has recently shown great promise in computer vision, but it has mostly
focused on synthesizing visually realistic images. In this paper, motivated by multi-task …

[HTML][HTML] Few-Shot Learning for Clinical Natural Language Processing Using Siamese Neural Networks: Algorithm Development and Validation Study

D Oniani, P Chandrasekar, S Sivarajkumar, Y Wang - JMIR AI, 2023 - ai.jmir.org
Background Natural language processing (NLP) has become an emerging technology in
health care that leverages a large amount of free-text data in electronic health records to …

An exploratory journey of representation learning's enhancement, adaptation and related intelligent methods

J Wu - 2024 - ideals.illinois.edu
Abstract Representation learning models employing Siamese structures have consistently
demonstrated exceptional performance across various fields, including deep learning …