Chang Che
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Chang Che has a total of 13 co-authors. Here are the top co-authors by number of shared publications:
- Dan Hu
Beijing Normal University
2 shared publications - Hamid Reza Karimi
Politecnico di Milano
1 shared publication - Jiaxin Huang
Trine University, USA
1 shared publication - Jin Wang1 shared publication
- Jiuwen Cao
Hangzhou Dianzi University
1 shared publication - Liqiang Yu
The University of Chicago, USA
1 shared publication
Latest Publications
- Top Cited Papers
- 29 January 2026
- Published by Elsevier in Information Fusion
- Vol. 131
Information-theoretic graph fusion with vision-language-action model for policy reasoning and dual robotic control
- 29 January 2026
- Published by Elsevier in Information Fusion
- Vol. 131
Top Cited PapersInformation-theoretic graph fusion with vision-language-action model for policy reasoning and dual robotic control
- Open Access
- 22 September 2023
- Published by Association for Computing Machinery (ACM)
- p. 414-418
Enhancing Multimodal Understanding with CLIP-Based Image-to-Text Transformation
The process of transforming input images into corresponding textual explanations stands as a crucial and complex endeavor within the domains of computer vision and natural language processing. In this paper, we propose an innovative ensemble approach that harnesses the capabilities of Contrastive Language-Image Pretraining (CLIP) models. Our ensemble framework encompasses two significant variations of the CLIP model, each meticulously designed to cater to specific nuances within the image-to-text transformation landscape. The first model introduces an elaborated architecture, featuring multiple layers with distinct learning rates, thereby amplifying its adeptness in capturing intricate relationships between images and text. The second model strategically exploits CLIP’s inherent zero-shot learning potential to generate image-text embeddings, subsequently harnessed by a K-Nearest Neighbors (KNN) model. Through this KNN-based paradigm, the model facilitates image-to-text transformation by identifying closely related embeddings within the embedding space. Notably, our ensemble approach is rigorously evaluated, employing the cosine similarity metric to gauge the alignment between model-generated embeddings and ground truth representations. Comparative experiments vividly highlight the superiority of our ensemble strategy over standalone CLIP models. This study not only advances the state-of-the-art in image-to-text transformation but also accentuates the promising trajectory of ensemble learning in effectively addressing intricate multimodal tasks.
- 22 September 2023
- Published by Association for Computing Machinery (ACM)
- p. 414-418
Open AccessEnhancing Multimodal Understanding with CLIP-Based Image-to-Text Transformation
The process of transforming input images into corresponding textual explanations stands as a crucial and complex endeavor within the domains of computer vision and natural language processing. In this paper, we propose an innovative ensemble approach that harnesses the capabilities of Contrastive Language-Image Pretraining (CLIP) models. Our ensemble framework encompasses two significant variations of the CLIP model, each meticulously designed to cater to specific nuances within the image-to-text transformation landscape. The first model introduces an elaborated architecture, featuring multiple layers with distinct learning rates, thereby amplifying its adeptness in capturing intricate relationships between images and text. The second model strategically exploits CLIP’s inherent zero-shot learning potential to generate image-text embeddings, subsequently harnessed by a K-Nearest Neighbors (KNN) model. Through this KNN-based paradigm, the model facilitates image-to-text transformation by identifying closely related embeddings within the embedding space. Notably, our ensemble approach is rigorously evaluated, employing the cosine similarity metric to gauge the alignment between model-generated embeddings and ground truth representations. Comparative experiments vividly highlight the superiority of our ensemble strategy over standalone CLIP models. This study not only advances the state-of-the-art in image-to-text transformation but also accentuates the promising trajectory of ensemble learning in effectively addressing intricate multimodal tasks.
- 1 January 2010
- Published by Springer Nature
- p. 811-817
An Approach Based on Hilbert-Huang Transform and Support Vector Machine for Intelligent Fault Diagnosis
- 1 January 2010
- Published by Springer Nature
- p. 811-817
An Approach Based on Hilbert-Huang Transform and Support Vector Machine for Intelligent Fault Diagnosis
- 1 January 2008
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 2 (21579555) , 183-187
Support Vector Reproducing Kernel Based on Walsh Series for Regression
A new reproducing kernel function of least square support vector machine (SVM) based on Walsh series is presented in this paper. The reproducing kernel has been constructed in reproducing kernel Hilbert space (RKHS). Because the Hilbert space and the square integrable space are isomorphic, according to the wavelet multi-resolution analysis, Walsh consequence can be seen a set of orthogonal basis to construct the reproducing kernel. The simulation results are discussed to illustrate the proposed reproducing kernel.
- 1 January 2008
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- Vol. 2 (21579555) , 183-187
Support Vector Reproducing Kernel Based on Walsh Series for Regression
A new reproducing kernel function of least square support vector machine (SVM) based on Walsh series is presented in this paper. The reproducing kernel has been constructed in reproducing kernel Hilbert space (RKHS). Because the Hilbert space and the square integrable space are isomorphic, according to the wavelet multi-resolution analysis, Walsh consequence can be seen a set of orthogonal basis to construct the reproducing kernel. The simulation results are discussed to illustrate the proposed reproducing kernel.