Dexiang Deng

Papers

Total Citations

H-Index

About

Dexiang Deng is a researcher whose work sits at the intersection of computer vision and deep learning, with a particular focus on face analysis and object recognition. His most impactful contribution, "Face age classification based on a deep hybrid model" (2018), has garnered 14 citations, demonstrating his ability to combine neural network architectures for fine-grained visual tasks. In this work, Deng proposed a hybrid deep learning framework that effectively fuses features for age estimation from facial images—a challenging problem with applications in security, marketing, and human-computer interaction. Earlier, in "An improved real-time object proposals generation method based on local binary pattern" (2017), he tackled the critical preprocessing step of object detection by developing a method that generates category-independent object proposals in real time. By integrating Local Binary Pattern features into the proposal pipeline, Deng achieved faster and more efficient object localization, offering a practical alternative to traditional sliding window approaches. Though his citation counts are modest, Deng’s research reflects a consistent focus on practical, real-time vision systems—bridging classical feature engineering with modern deep learning to solve applied problems in image understanding.