Jinqing Zheng

Jinqing Zheng

Auto-generated profile, based on publicly available information

Jinqing Zheng

College of Intelligence and Computing, Tianjin University, Tianjin, China

Research Focus & Metrics

Academic Career Length

11Years (2016-Present)

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

Representative Publications

View All Publications
5 April 2023 in IEEE Transactions on Multimedia
Verified
Journal Article
Closed access
CQCheng Qi
Cheng Qi
ZFZhiyong Feng
Zhiyong Feng
MXMeng Xing
Meng Xing
YSYong Su
Yong Su
JZJinqing Zheng
Jinqing Zheng
YZYiming Zhang
Yiming Zhang
Recently, Action Recognition (AR) is facing the scalability problem, since collecting and annotating data for the ever-growing action categories is exhausting and inappropriate. As an alternative to AR, Zero-Shot Action Recognition (ZSAR) is getting more and more attention in the community, as they could utilize a shared semantic/attribute space to recognize novel categories without annotated data. Different from the AR focuses on learning the correlation between actions, ZSAR needs to consider the correlation of action-action, label-label and action-label at the same time. However, as far as we know, there is no work to provide structural guidance for the framework design of ZSAR according to its task characteristics. In this paper, we demonstrate the rationality of using the Energy-Based Model (EBM) to guide the framework design of ZSAR based on their inference mechanism. Furthermore, under the guidance of EBM, we propose an Energy-based Temporal Summarized Attentive Network (ETSAN) to achieve ZSAR. Specifically, to ensure the effectiveness of cross-modal matching, EBM needs to capture the correlations of input-input, output-output and input-output, based on discriminative and focused input and output space. To this end, we first design the Temporal Summarized Attentive Mechanism (TSAM) to capture the correlation of action-action by constructing discriminative and focused input space. Then, a Label Semantic Adaptive Mechanism (LSAM) is proposed to learn the correlation of label-label by adjusting the semantic structure according to the target task. Finally, we devise an Energy Score Estimation Mechanism (ESEM) to measure the compatibility (i.e. energy score) between video representation and label semantic embedding. With end-to-end training, our framework can capture all three of the correlations mentioned above simultaneously by minimizing the energy score of the correct action-label pair. Experiments on the HMDB51 and UCF101 datasets show that the proposed architecture achieves comparable results among methods based on the spatial-temporal visual feature of sequence-level, which demonstrates the efficiency of the EBM in guiding the framework design of ZSAR. In addition, our code is available at https://github.com/oOHCIOo/ETSAN .

28 September 2016 in Multimedia Tools and Applications
Verified
Journal Article
Closed access
PDF file
JZJinqing Zheng
Jinqing Zheng
ZFZhiyong Feng
Zhiyong Feng
CXChao Xu
Chao Xu
JHJing Hu
Jing Hu
WGWeimin Ge
Weimin Ge

3 February 2026 in IEEE Transactions on Geoscience and Remote Sensing
Verified
Journal Article
Closed access
JZJinqing Zheng
Jinqing Zheng
ZFZhiyong Feng
Zhiyong Feng
MXMeng Xing
Meng Xing
YSYong Su
Yong Su
MFMeie Fang
Meie Fang
YZYiming Zhang
Yiming Zhang
Oriented Vehicle Detection (OVD) in drone imagery plays a fundamental role in a wide range of downstream applications, owing to its capacity to provide accurate localization and corresponding categorization of vehicle instances in complex scenes. Existing OVD methods for drone-based scenes are limited by their coarse granularity, while real-world applications often require fine-grained detection—for example, distinguishing between van trucks and tank trucks is critical for traffic regulation and road safety. To address this gap, we introduce Fine-Grained Oriented Vehicle Detection (FGOVD) task for the first time in drone-based scenarios. To facilitate this task, we develop a Semi-Automatic Labeling approach for Fine-Grained Oriented Object Detection, named SAL-FGOOD, and construct a truly fine-grained multi-scene dataset, Drone Fine-Grained Oriented Vehicle Detection (DFGOVD) dataset. SALFGOOD extends the existing labeling paradigm by incorporating a category-adjustment branch, reducing manual workload by 94.7% for our DFGOVD dataset and 77.7% for the public FAIR1M-v2 dataset. Building upon SAL-FGOOD, we construct DFGOVD, the first large-scale and comprehensive FGOVD dataset for drone-based scenes. The dataset comprises 33,669 images and 816,239 vehicle instances spanning 53 categories and captures subtle distinctions in vehicle appearance, size, and functionality. We benchmark state-of-the-art oriented object detection methods on the DFGOVD dataset, exposing its inherent complexity and challenge. The dataset is available upon request from https://github.com/JinqingZhengTju/DFGOVD.

28 January 2024
Verified
Book Chapter
Closed access
PDF file
PCPei Chen
Pei Chen
ZFZhiyong Feng
Zhiyong Feng
MXMeng Xing
Meng Xing
YZYiming Zhang
Yiming Zhang
JZJinqing Zheng
Jinqing Zheng

20 August 2026 in SSRN Electronic Journal
Verified
Preprint
Open access
MXMeng Xing
Meng Xing
YSYong Su
Yong Su
QPQifa Peng
Qifa Peng
JZJinqing Zheng
Jinqing Zheng
SZShuang Zhu
Shuang Zhu
YXYu Xing
Yu Xing
Human Pose Estimation (HPE) is a core problem in computer vision, with applications in human-computer interaction, sports analytics, and augmented reality. However, the collection and use of data for deploying existing HPE methods typically involve biometric cues, such as facial features and body shape, potentially exposing these methods to regulatory risks under the EU General Data Protection Regulation (GDPR) and the Illinois Biometric Information Privacy Act (BIPA).In this paper, we introduce LHPE, a legal-compliant HPE setting where pose estimation is performed from reflections on transparent surfaces, naturally limiting access to low-level visual cues (e.g., texture and color) that may expose identifiable biometric information.This setting introduces two key challenges: (1) reflection-induced distortions and incomplete visual information hinder reliable annotation, and (2) the absence of low-level cues degrades pose estimation performance. To address the former, we construct the Transparent Medium-based Reflection Image (TMRI) dataset with motion-capture-based annotations, ensuring high-quality labels despite severe visual ambiguity. To address the latter, we propose ReflecPoseNet, a dual-stream architecture tailored for robust pose estimation under reflection-only observations.Extensive experiments demonstrate that our approach achieves competitive performance on real reflection data, validating both the TMRI dataset and the proposed legal-compliant HPE paradigm. Code and dataset are available at https://github.com/oOHCIOo/Legal-Compliant-Human-Pose-Estimation.