Author name cluster

Chenqiang Gao

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

3 papers

1 author row

EAAI Journal 2026 Journal Article

Ground-to-Aerial Scene Adaptation: Unsupervised drone video action recognition via domain adaptation

Feng Yang
Zhijia Li
Xi Liu
Fulin Luo
Anyong Qin
Tiecheng Song
Yue Zhao
Chenqiang Gao

The unsupervised domain adaptation (UDA) methods can well address the problem of scarce annotated videos in drone video action recognition by transferring knowledge from rich ground-camera datasets. However, existing UDA video methods are subject to huge domain shifts: changes in perspective and scale for key areas result in significant differences between ground-camera videos and drone videos, posing distinct alignment challenges. Moreover, the high similarity between classes in drone videos further complicates the learning of class-discriminant information. To deal with these issues, we propose a novel framework called Ground-to-Aerial Scene Adaptation (GASA). To enhance the learning of class-discriminant information, we propose a View-Independent Attention (VIA) module that learns view-independent attention maps and generates discriminative features of action in different domains. Then, a contrastive loss is used to pull the samples from the same category closer in the feature space for semantic alignment, improving the discrimination of target domain classes. Furthermore, we design a novel relation-weighted alignment loss to learn domain-invariant information and focus on challenging samples for better domain alignment. We report the results on Kinetics → NEC-Drone benchmark and Kinetics → UAV-Human dataset, which illustrates the superior performance of our proposed approach. The code is available at https: //github. com/lzjjia/GASA.

Details DOI

EAAI Journal 2025 Journal Article

Frequency-prompt guided spectral–spatial transformer for hyperspectral image classification

Tiecheng Song
Lei Mao
Longlong Zhang
Anyong Qin
Feng Yang
Chenqiang Gao

Convolutional neural networks (CNNs) and vision transformers (ViTs) have been applied to hyperspectral image (HSI) classification. However, CNNs have difficulty in capturing global context information, while ViTs lack the excellent ability of CNNs to capture local relationships in images. How to integrate CNNs and ViTs to design an effective network remains a challenge. In addition, most of the HSI classification methods are built upon the original image domain, failing to utilize the prior knowledge of the frequency domain. Although some studies explore the frequency-domain information, they cannot effectively achieve spatial-spectral-frequency information interactions and deep feature fusion. To solve the above challenges, we propose a frequency-prompt guided spectral–spatial transformer (FPGSST) for HSI classification. The backbone of FPGSST adopts a two-branch hierarchical network structure, where each branch extracts local and global spatial/spectral feature based on multi-stage convolutions and transformers. To guide the network to learn discriminative spatial and spectral features, we carefully design spatial-wise and spectral-wise frequency prompt modules. These prompts are structural which contain low-frequency priors and learnable high-frequency components. By embedding them into the self-attention mechanism, our model can focus on frequency-domain information in both spatial and spectral domains. We also design a spectral–spatial fusion transformer and a multi-stage fusion transformer to integrate comprehensive features. The resulting FPGSST achieves local–global information interaction and multi-stage spectral–spatial-frequency feature fusion by flexibly harnessing the variants of transformer architecture. Experimental results on several datasets demonstrate the superiority of FPGSST over the compared methods for HSI classification.

Details DOI

AAAI Conference 2016 Conference Paper

Two-Stream Contextualized CNN for Fine-Grained Image Classification

Jiang Liu
Chenqiang Gao
Deyu Meng
Wangmeng Zuo

Human’s cognition system prompts that context information provides potentially powerful clue while recognizing objects. However, for ﬁne-grained image classiﬁcation, the contribution of context may vary over different images, and sometimes the context even confuses the classiﬁcation result. To alleviate this problem, in our work, we develop a novel approach, two-stream contextualized Convolutional Neural Network, which provides a simple but efﬁcient contextcontent joint classiﬁcation model under deep learning framework. The network merely requires the raw image and a coarse segmentation as input to extract both content and context features without need of human interaction. Moreover, our network adopts a weighted fusion scheme to combine the content and the context classiﬁers, while a subnetwork is introduced to adaptively determine the weight for each image. According to our experiments on public datasets, our approach achieves considerable high recognition accuracy without any tedious human’s involvements, as compared with the state-of-the-art approaches.

PDF Details