Author name cluster

Jianzhong Wang

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 2025 Journal Article

A fine-grained detection algorithm for identity and actions of weapon-holding targets in public safety

Zibo Yu
Weichao Wu
Jianzhong Wang
Yu You
Shaobo Bian
Endi Wang

Fine-grained detection of threat targets in public security can provide more precise and detailed security information. Compared with general security systems, it can achieve faster and more accurate responses, which is of great significance for ensuring public security. In the face of the problems of low detection accuracy, inaccurate classification, and complex scene in fine-grained threat detection, we propose a fine-grained threat detection model based on the improved RealTime Detection Transformers (RT-DETR) algorithm. First, we built a high-performance backbone network called the Dilated Feature Aggregation Network to efficiently extract multiscale features. We used dilated convolutions for spatial feature aggregation and reparameterized them as large kernel convolutions to efficiently capture fine-grained target features. Secondly, a new feature encoder was developed to enable the model to better capture local and global features in complex real-world scenarios. Meanwhile, we propose a new feature fusion module DyContext Fusion Module, which improves the efficiency and fine-grained feature representation of feature fusion by dynamic sampling, context-guided features, and convolutional fusion. Finally, we utilized Adaptive Classification Loss to balance positive and negative samples, increased the inter-class diversity, and improved the model’s classification accuracy. We have built a public safety dataset and conducted comparative experiments with state-of-the-art algorithms that are publicly available to verify the effectiveness of the proposed improvement method. We also tested the performance of this method in terms of fine-grained and action detection on the Stanford 40 actions dataset, and the results showed that its performance was good. This proves the possibility of using this method for fine-grained detection of public security threat targets. The relevant code is released at https: //github. com/ZIXUAN98/FG-DETR.

Details DOI

AAAI Conference 2025 Conference Paper

Temporal Action Localization with Cross Layer Task Decoupling and Refinement

Qiang Li
Di Liu
Jun Kong
Sen Li
Hui Xu
Jianzhong Wang

Temporal action localization (TAL) involves dual tasks to classify and localize actions within untrimmed videos. However, the two tasks often have conflicting requirements for features. Existing methods typically employ separate heads for classification and localization tasks but share the same input feature, leading to suboptimal performance. To address this issue, we propose a novel TAL method with Cross Layer Task Decoupling and Refinement (CLTDR). Based on the feature pyramid of video, CLTDR strategy integrates semantically strong features from higher pyramid layers and detailed boundary-aware boundary features from lower pyramid layers to effectively disentangle the action classification and localization tasks. Moreover, the multiple features from cross layers are also employed to refine and align the disentangled classification and regression results. At last, a lightweight Gated Multi-Granularity (GMG) module is proposed to comprehensively extract and aggregate video features at instant, local, and global temporal granularities. Benefiting from the CLTDR and GMG modules, our method achieves state-of-the-art performance on five challenging benchmarks: THUMOS14, MultiTHUMOS, EPIC-KITCHENS-100, ActivityNet-1.3, and HACS. Code：https://github.com/LiQiang0307/CLTDR-GMG

PDF Details DOI

IJCAI Conference 2020 Conference Paper

Real-World Automatic Makeup via Identity Preservation Makeup Net

Zhikun Huang
Zhedong Zheng
Chenggang Yan
Hongtao Xie
Yaoqi Sun
Jianzhong Wang
Jiyong Zhang

This paper focuses on the real-world automatic makeup problem. Given one non-makeup target image and one reference image, the automatic makeup is to generate one face image, which maintains the original identity with the makeup style in the reference image. In the real-world scenario, face makeup task demands a robust system against the environmental variants. The two main challenges in real-world face makeup could be summarized as follow: first, the background in real-world images is complicated. The previous methods are prone to change the style of background as well; second, the foreground faces are also easy to be affected. For instance, the ``heavy'' makeup may lose the discriminative information of the original identity. To address these two challenges, we introduce a new makeup model, called Identity Preservation Makeup Net (IPM-Net), which preserves not only the background but the critical patterns of the original identity. Specifically, we disentangle the face images to two different information codes, i. e. , identity content code and makeup style code. When inference, we only need to change the makeup style code to generate various makeup images of the target person. In the experiment, we show the proposed method achieves not only better accuracy in both realism (FID) and diversity (LPIPS) in the test set, but also works well on the real-world images collected from the Internet.

PDF Details DOI