Zhimao Lai

Dr. Zhimao Lai

China people's police university, Guangzhou, Guangdong, China
orcid

Research Focus & Metrics

Academic Career Length

13Years (2014-Present)

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

Representative Publications

View All Publications
14 November 2024 in Electronics
Verified
Journal Article
Open access
PDF file
ZLZhimao Lai
Zhimao Lai
JLJicheng Li
Jicheng Li
CWChuntao Wang
Chuntao Wang
Jianhua WuJianhua Wu
Jianhua Wu
DJdonghua jiang
donghua jiang
The proliferation of AI-generated content (AIGC) has empowered non-experts to create highly realistic Deepfake images and videos using user-friendly software, posing significant challenges to the legal system, particularly in criminal investigations, court proceedings, and accident analyses. The absence of reliable Deepfake verification methods threatens the integrity of legal processes. In response, researchers have explored deep forgery detection, proposing various forensic techniques. However, the swift evolution of deep forgery creation and the limited generalizability of current detection methods impede practical application. We introduce a new deep forgery detection method that utilizes image decomposition and lighting inconsistency. By exploiting inherent discrepancies in imaging environments between genuine and fabricated images, this method extracts robust lighting cues and mitigates disturbances from environmental factors, revealing deeper-level alterations. A crucial element is the lighting information feature extractor, designed according to color constancy principles, to identify inconsistencies in lighting conditions. To address lighting variations, we employ a face material feature extractor using Pattern of Local Gravitational Force (PLGF), which selectively processes image patterns with defined convolutional masks to isolate and focus on reflectance coefficients, rich in textural details essential for forgery detection. Utilizing the Lambertian lighting model, we generate lighting direction vectors across frames to provide temporal context for detection. This framework processes RGB images, face reflectance maps, lighting features, and lighting direction vectors as multi-channel inputs, applying a cross-attention mechanism at the feature level to enhance detection accuracy and adaptability. Experimental results show that our proposed method performs exceptionally well and is widely applicable across multiple datasets, underscoring its importance in advancing deep forgery detection.

7 March 2022 in Systems
Verified
Journal Article
Open access
PDF file
ZLZhimao Lai
Zhimao Lai
YWYufei Wang
Yufei Wang
Renhai FengRenhai Feng
Renhai Feng
XHXianglei Hu
Xianglei Hu
HXhaifeng Xu
haifeng Xu
With the rapid development of deep learning, generating realistic fake face videos is becoming easier. It is common to make fake news, network pornography, extortion and other related illegal events using deep forgery. In order to attenuate the harm of deep forgery face video, researchers proposed many detection methods based on the tampering traces introduced by deep forgery. However, these methods generally have poor cross-database detection performance. Therefore, this paper proposes a multi-feature fusion detection method to improve the generalization ability of the detector. This method combines feature information of face video in the spatial domain, frequency domain, Pattern of Local Gravitational Force (PLGF) and time domain and effectively reduces the average error rate of span detection while ensuring good detection effect in the library.

16 December 2024 in Electronics
Verified
Journal Article
Open access
PDF file
ZLZhimao Lai
Zhimao Lai
ZYZhuangxi Yao
Zhuangxi Yao
GLGuanyu Lai
Guanyu Lai
CWChuntao Wang
Chuntao Wang
Renhai FengRenhai Feng
Renhai Feng
The rapid advancement of Artificial Intelligence Generated Content (AIGC) has significantly accelerated the evolution of Deepfake technology, thereby introducing escalating social risks due to its potential misuse. In response to these adverse effects, researchers have developed defensive measures, including passive detection and proactive forensics. Although passive detection has achieved some success in identifying Deepfakes, it encounters challenges such as poor generalization and decreased accuracy, particularly when confronted with anti-forensic techniques and adversarial noise. As a result, proactive forensics, which offers a more resilient defense mechanism, has garnered considerable scholarly interest. However, existing proactive forensic methodologies often fall short in terms of visual quality, detection accuracy, and robustness. To address these deficiencies, we propose a novel proactive forensic approach that utilizes pseudo-Zernike moment robust watermarking. This method is specifically designed to enhance the detection and analysis of face swapping by transforming facial data into a binary bit stream and embedding this information within the non-facial regions of video frames. Our approach facilitates the detection of Deepfakes while preserving the visual integrity of the video content. Comprehensive experimental evaluations have demonstrated the robustness of this method against standard signal processing operations and its superior performance in detecting Deepfake manipulations.

7 October 2022 in Applied Optics
Verified
Journal Article
Closed access
PDF file
Renhai FengRenhai Feng
Renhai Feng
ZZZhaolin Zhang
Zhaolin Zhang
ZLZhimao Lai
Zhimao Lai
XMXurui Mao
Xurui Mao
In view of the complexity of port ship supervision and the influence of external factors such as electromagnetic interference in harbor-border inspection, an efficient system combining an unmanned aerial vehicle (UAV) and visible light positioning (VLP) is proposed for locating maritime targets. In this system, a rotatable receiver with five photodetectors (PDs) installed obliquely on UAV is designed for expanding the positioning range and allowing a lower flight altitude. On this basis, we propose the Chan–Taylor (CT) method based on time difference of arrival (TDOA) for target positioning. First, the localization problem is reformulated as a weighted least squares (WLS) problem and provides a good initial estimate via the two-step WLS (TWLS) method. Then, based on Taylor expansion of TDOA equations, estimated error is calculated using the initial estimate, which can correct the estimated position of the target iteratively. To offset the error, weighted centroid CT (WCCT) is proposed by endowing different weights based on error difference to estimated results. For further improving accuracy, a restricted-region fingerprinting positioning based on CT (CT-RFP) is proposed. In restricted area determined by CT, a certain number of fingerprints is generated based on received signal strength (RSS) for matching. Simulation results show that CT is significantly improved over the previous methods. Compared with TWLS, the accuracy of CT is improved by 49.71%. For WCCT, the maximum error is reduced from 8.65 to 6.91 cm, which effectively reduces the influence of error. Moreover, CT-RFP can achieve an accuracy within millimeter level via the appropriate number of fingerprints and ensemble runs of CT, even at high noise levels.

22 May 2014 in Multimedia Tools and Applications
Verified
Journal Article
Open access
PDF file
YHYongjian Hu
Yongjian Hu
CLChang-Tsun Li
Chang-Tsun Li
ZLZhimao Lai
Zhimao Lai
Fast camera fingerprint search is an important issue for source camera identification in real-world applications. So far there has been little work done in this area. In this paper, we propose a novel fast search algorithm. We use global information derived from the relationship between the query fingerprint/digest and the reference fingerprints/digests in the database to guide fast search. This information can provide more accurate and robust clues for the selection of candidate matching database fingerprints. Because the quality of query fingerprints may degrade or vary in realistic applications, the construction of robust search clues is significant. To speed up the search process, we adopt a lookup table that is built on the separate-chaining hash table. The proposed algorithm has been tested using query images from real-world photos. Experiments demonstrate that our algorithm can well adapt to query fingerprints with different quality. It can achieve higher detection rates with lower computational cost than the traditional brute-force search algorithm and a pioneering fast search algorithm in literature.

23 October 2024 in Sensors
Verified
Journal Article
Open access
PDF file
ZLZhimao Lai
Zhimao Lai
YZYan Zhang
Yan Zhang
ALAnd Xiubo Liang
And Xiubo Liang
Human action recognition (HAR) is a critical area in computer vision with wide-ranging applications, including video surveillance, healthcare monitoring, and abnormal behavior detection. Current HAR methods predominantly rely on full-body data, which can limit their effectiveness in real-world scenarios where occlusion is common. In such situations, the face often remains visible, providing valuable cues for action recognition. This paper introduces Face in Action (FIA), a novel two-stream method that leverages facial action cues for robust action recognition under conditions of significant occlusion. FIA consists of an RGB stream and a landmark stream. The RGB stream processes facial image sequences using a fine-spatio-multitemporal (FSM) 3D convolution module, which employs smaller spatial receptive fields to capture detailed local facial movements and larger temporal receptive fields to model broader temporal dynamics. The landmark stream processes facial landmark sequences using a normalized temporal attention (NTA) module within an NTA-GCN block, enhancing the detection of key facial frames and improving overall recognition accuracy. We validate the effectiveness of FIA using the NTU RGB+D and NTU RGB+D 120 datasets, focusing on action categories related to medical conditions. Our experiments demonstrate that FIA significantly outperforms existing methods in scenarios with extensive occlusion, highlighting its potential for practical applications in surveillance and healthcare settings.