Zhi Li

Dr. Zhi Li


Research Focus & Metrics

Academic Career Length

6Years (2021-Present)

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

Representative Publications

View All Publications
1 May 2023 in IEEE Journal of Biomedical and Health Informatics
Verified
Journal Article
Closed access
QJQunchao Jin
Qunchao Jin
HHHongyu Hou
Hongyu Hou
GZGuixu Zhang
Guixu Zhang
ZLZhi Li
Zhi Li
Regular colonoscopy is an effective way to prevent colorectal cancer by detecting colorectal polyps. Automatic polyp segmentation significantly aids clinicians in precisely locating polyp areas for further diagnosis. However, polyp segmentation is a challenge problem, since polyps appear in a variety of shapes, sizes and textures, and they tend to have ambiguous boundaries. In this paper, we propose a U-shaped model named Feedback Enhancement Gate Network (FEGNet) for accurate polyp segmentation to overcome these difficulties. Specifically, for the high-level features, we design a novel Recurrent Gate Module (RGM) based on the feedback mechanism, which can refine attention maps without any additional parameters. RGM consists of Feature Aggregation Attention Gate (FAAG) and Multi-Scale Module (MSM). FAAG can aggregate context and feedback information, and MSM is applied for capturing multi-scale information, which is critical for the segmentation task. In addition, we propose a straightforward but effective edge extraction module to detect boundaries of polyps for low-level features, which is used to guide the training of early features. In our experiments, quantitative and qualitative evaluations show that the proposed FEGNet has achieved the best results in polyp segmentation compared to other state-of-the-art models on five colonoscopy datasets.

28 August 2024 in IEEE Transactions on Circuits and Systems for Video Technology
Verified
Journal Article
Closed access
BCBodong Cheng
Bodong Cheng
Juncheng LiJuncheng Li
Juncheng Li
JSJun Shi
Jun Shi
YFYingying Fang
Yingying Fang
GZGuixu Zhang
Guixu Zhang
ZLZhi Li
Zhi Li
The extraction of distribution from images with diverse weather conditions is crucial for enhancing the robustness of visual algorithms. When addressing image degradation caused by different weather, accurately perceiving the data distribution of weather-informed degradation becomes a fundamental challenge. However, given the highly stochastic nature, modelling weather distribution poses a formidable task. In this paper, we propose a novel multi-Weather distribution difFUsion blind restoration model, named WeaFU. Firstly, the model employs representation learning to map image distribution into a latent space. Subsequently, WeaFU utilizes a diffusion-based approach, with the assistance of Diffusion Distribution Generator (DDG), to perceive and extract corresponding weather distribution. This strategy ingeniously injects data distribution into the recovery process, significantly enhancing the robustness of the model in diverse weather scenarios. Finally, a Conditional Distribution-Aware Transformer (CDAT) is constructed to align the distribution information with pixels, thereby obtaining clear images. Extensive experiments on real and synthetic datasets demonstrate that WeaFU achieves superior performance.

22 January 2021 in IET Image Processing
Verified
Journal Article
Open access
PDF file
TWTingting Wu
Tingting Wu
XGXiaoyu Gu
Xiaoyu Gu
JSJinbo Shao
Jinbo Shao
RZRuoxuan Zhou
Ruoxuan Zhou
ZLZhi Li
Zhi Li
Image segmentation is a fundamental and challenging task in image processing and computer vision. The colour image segmentation is attracting more attention as the colour image provides more information than the grey image. A variational model based on a convex K‐means approach to segment colour images is proposed. The proposed variational method uses a combination of l 1 and l 2 regularizers to maintain edge information of objects in images while overcoming the staircase effect. Meanwhile, our one‐stage strategy is an improved version based on the smoothing and thresholding strategy, which contributes to improving the accuracy of segmentation. The proposed method performs the following steps. First, the colour set which can be determined by human or the K‐means method is specified. Second, a variational model to obtain the most appropriate colour for each pixel from the colour set via convex relaxation and lifting is used. The Chambolle–Pock algorithm and simplex projection are applied to solve the variational model effectively. Experimental results and comparison analysis demonstrate the effectiveness and robustness of the method.

25 December 2024 in IEEE Transactions on Multimedia
Verified
Journal Article
Closed access
MQMengzan Qi
Mengzan Qi
SCSixian Chan
Sixian Chan
CHChen Hang
Chen Hang
GZGuixu Zhang
Guixu Zhang
TZTieyong ZENG
Tieyong ZENG
ZLZhi Li
Zhi Li
Visible-Infrared Person Re-identification aims to retrieve images of specific identities across modalities. To relieve the large cross-modality discrepancy, researchers introduce the auxiliary modality within the image space to assist modalityinvariant representation learning. However, the challenge persists in constraining the inherent quality of generated auxiliary images, further leading to a bottleneck in retrieval performance. In this paper, we propose a novel Auxiliary Representation Guided Network (ARGN) to explore the potential of auxiliary representations, which are directly generated within the modalityshared embedding space. In contrast to the original visible and infrared representations, which contain information solely from their respective modalities, these auxiliary representations integrate cross-modality information by fusing both modalities. In our framework, we utilize these auxiliary representations as modality guidance to reduce the cross-modality discrepancy. Firstly, we propose a High-quality Auxiliary Representation Learning (HARL) framework to generate identity-consistent auxiliary representations. The primary objective of our HARL is to ensure that auxiliary representations capture diverse modality information from both modalities while concurrently preserving identity-related discrimination. Secondly, guided by these auxiliary representations, we design an Auxiliary Representation Guided Constraint (ARGC) to optimize the modality-shared embedding space. By incorporating this constraint, the modalityshared embedding space is optimized to achieve enhanced intraidentity compactness and inter-identity separability, further improving the retrieval performance. In addition, to improve the robustness of our framework against the modality variation, we introduce a Part-based Adaptive Gaussian Module (PAGM) to adaptively extract discriminative information across modalities. Finally, extensive experiments are conducted to demonstrate the superiority of our method over state-of-the-art approaches on three VI-ReID datasets.

5 August 2023 in Signal Processing
Verified
Journal Article
Closed access
FLFang Li
Fang Li
FFFaming Fang
Faming Fang
ZLZhi Li
Zhi Li
TZTieyong ZENG
Tieyong ZENG

28 October 2024
Verified
Conference Paper
Open access
SJShilong Jia
Shilong Jia
TWTingting Wu
Tingting Wu
YFYingying Fang
Yingying Fang
TZTieyong ZENG
Tieyong ZENG
GZGuixu Zhang
Guixu Zhang
ZLZhi Li
Zhi Li
Incremental Object Detection (IOD) simulates the dynamic data flow in real-world applications, which require detectors to learn new classes or adapt to new domains while retaining knowledge from previous tasks. Most existing IOD methods focus only on class incremental learning, assuming all data comes from the same domain. However, this is hardly achievable in practical applications, as images collected under different conditions often exhibit completely different characteristics, such as lighting, weather, style, etc. Class IOD methods suffer from performance degradation in these scenarios with domain shifts. To bridge domain shifts and category gaps in IOD, we propose Purified Distillation (PD), where we use a set of trainable queries to transfer the teacher's attention on old tasks to the student and adopt the gradient reversal layer to guide the student to learn the teacher's feature space structure from a micro perspective, which has not been extensively studied in previous works. Meanwhile, PD combines classification confidence with localization confidence to purify the most meaningful output nodes, so that the student model inherits a more comprehensive teacher knowledge. Extensive experiments across various IOD settings on six widely used datasets show that PD significantly outperforms state-of-the-art methods. Even after five steps of incremental learning, our method can preserve 60.6% mAP on the first task, while compared methods can only maintain up to 55.9%.