The capability of the traditional semi-supervised learning (SSL) methods is far from real-
world application due to severely biased pseudo-labels caused by (1) class imbalance and …

Recent state-of-the-art active learning methods have mostly leveraged generative
adversarial networks (GANs) for sample acquisition; however, GAN is usually known to …

Abstract The task of Visual Question Answering (VQA) is known to be plagued by the issue
of VQA models exploiting biases within the dataset to make its final prediction. Various …

Visible-infrared person re-identification (ReID) aims to recognize a same person of interest
across a network of RGB and IR cameras. Some deep learning (DL) models have directly …

Contrastive language-image pretraining (CLIP) links vision and language modalities into a
unified embedding space, yielding the tremendous potential for vision-language (VL) tasks …

We introduce dense relational captioning, a novel image captioning task which aims to
generate multiple captions with respect to relational information between objects in a visual …

In medical vision, different imaging modalities provide complementary information. However,
in practice, not all modalities may be available during inference or even training. Previous …

A common problem in the task of human-object interaction (HOI) detection is that numerous
HOI classes have only a small number of labeled examples, resulting in training sets with a …

The goal of this work is background-robust continuous sign language recognition. Most
existing Continuous Sign Language Recognition (CSLR) benchmarks have fixed …

Multimodal Sentiment Analysis (MSA) has attracted widespread research attention recently.
Most MSA studies are based on the assumption of signal completeness. However, many …