Changzhi Wang

Prof. Changzhi Wang

orcid

Research Focus & Metrics

Academic Career Length

5Years (2022-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
19 August 2022 in Expert Systems with Applications
Verified
Journal Article
Closed access
CWChangzhi Wang
Changzhi Wang
XGXiaodong Gu
Xiaodong Gu

14 April 2026 in IEEE Journal of Selected Topics in Applied Earth Observations and Remote Sensing
Verified
Journal Article
Open access
CWChangzhi Wang
Changzhi Wang
GZGuowei Zheng
Guowei Zheng
RZRenyue Zhang
Renyue Zhang
YZYuhan Zhu
Yuhan Zhu
FZFengjun Zhou
Fengjun Zhou
XLXing Liu
Xing Liu
Generating accurate and contextually coherent descriptions in remote sensing image captioning (RSIC) remains a formidable challenge, owing to the complex spatial structures of aerial and satellite imagery, as well as the requirement for high-level semantic interpretation. Conventional RSIC models predominantly rely on visual features alone, thereby overlooking rich linguistic priors that could significantly improve narrative consistency and scene understanding. To address this, we propose DPPF: Dual-Path Pre-Fusion with Semantic-Guided Encoding, a novel framework that integrates visual and semantic modalities at an early stage to leverage their complementary strengths. Specifically, we first employ the CLIP vision encoder to extract fine-grained patch-level visual features, which are fed as key-value pairs into the cross-attention layers of a transformer decoder, offering precise spatial grounding for word-level generation. Parallelly, we generate a scene-level caption using a pre-trained large language model (LLM) and encode it into a global semantic vector via the CLIP text encoder. Crucially, we further introduce a pre-fusion mechanism that concatenates this semantic vector with word embeddings to construct the input sequence for the decoder's masked self-attention, thereby allowing the language model to generate captions conditioned on holistic scene semantics from the outset. By separating visual and semantic encoding while injecting semantic context early, our method achieves balanced and effective multimodal interaction. Comprehensive experiments on three benchmark RSIC datasets show that DPPF achieves competitive performance compared to current state-of-the-art approaches, demonstrating the advantages of semantic-guided pre-fusion for remote sensing image captioning.

1 January 2026 in IEEE Geoscience and Remote Sensing Letters
Verified
Journal Article
Closed access
CWChangzhi Wang
Changzhi Wang
SJShaoze Jiao
Shaoze Jiao
YZYuhan Zhu
Yuhan Zhu
HGHaoran Guo
Haoran Guo
RWRui Wu
Rui Wu

1 November 2026 in IEEE Robotics and Automation Letters
Verified
Journal Article
Closed access
CWChangzhi Wang
Changzhi Wang
TYTinghao Yi
Tinghao Yi
ZHZhenzhen Huang
Zhenzhen Huang
ECEnhong Chen
Enhong Chen

27 October 2026 in Biomimetics
Verified
Journal Article
Open access
PDF file
QZQi Zhang
Qi Zhang
ZFZhiquan Feng
Zhiquan Feng
XKXueyu Kang
Xueyu Kang
HFHailing Feng
Hailing Feng
CWChangzhi Wang
Changzhi Wang
XYXinyu Yang
Xinyu Yang
Human action understanding aims to detect action instances and classify them in videos. Existing methods capture contextual dependencies through temporal feature aggregation or attention-based interactions. However, these methods primarily rely on visual features to establish relationships, making them susceptible to interference from background content or irrelevant action frames. To address this limitation, we propose Hybrid Causal Action Reasoning (HCAR), a framework for modeling causal correlations between action proposals. Specifically, we design a Hybrid Causality-Aware Attention (HCAA) to refine the semantic representations of proposal classes. To further characterize relations between adjacent proposals, an ordered relation branch evaluates the refined representations in temporal order, while a hyperbolic branch measures hierarchical semantic relatedness between action categories in the Poincaré ball. The outputs of these branches are combined to obtain causal correlation scores, which are subsequently used to refine action proposals. Extensive experiments on THUMOS14 and ActivityNet-1.3 demonstrate the effectiveness of our HCAR.