Yiting Wang
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Research Subjects
Yiting Wang has a total of 39 co-authors. Here are the top co-authors by number of shared publications:
- Valentina Donzella
Queen Mary University of London, London, U.K.
6 shared publications - Haonan Zhao
WMG, University of Warwick, U.K.
4 shared publications - Kurt Debattista
WMG, University of Warwick, Coventry, U.K.
4 shared publications Valentina Donzella
School of Engineering and Materials Science, Queen Mary University of London, U.K.
4 shared publications- Zixiang Wei
WMG, University of Warwick, Coventry, U.K.
4 shared publications - Lin Wang
Nanyang technological university, Singapore, Singapore
3 shared publications
Latest Publications
- Open Access
- 6 October 2026
- Published by Elsevier in Information Fusion
- Vol. 139
AURORA-KITTI: Any-weather depth completion and denoising
- 6 October 2026
- Published by Elsevier in Information Fusion
- Vol. 139
Open AccessAURORA-KITTI: Any-weather depth completion and denoising
- Open Access
- 2 March 2026
- Published by Elsevier in Knowledge-Based Systems
- Vol. 340
Rethinking probabilistic learning for counterfactual low-light image enhancement in robust engineering vision systems
- 2 March 2026
- Published by Elsevier in Knowledge-Based Systems
- Vol. 340
Open AccessRethinking probabilistic learning for counterfactual low-light image enhancement in robust engineering vision systems
- 19 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 26 (3) , 5013-5025
A Unified Generative Framework for Realistic LiDAR Simulation in Autonomous Driving Systems
Simulation models for perception sensors are integral components of automotive simulators used for the virtual validation of Autonomous Driving Systems (ADS). They also function as powerful tools for generating synthetic datasets to train deep learning-based perception models. LiDAR is a widely adopted perception sensor in ADS due to its high precision in 3D environment scanning. However, developing realistic LiDAR simulation models is a significant technical challenge. Unrealistic simulations can result in a large gap between the synthesised and real-world point clouds, limiting their effectiveness in perception and planning modules. Recently, deep generative models have emerged as promising solutions to synthesise realistic sensory data. However, for LiDAR simulation, deep generative models have been primarily hybridised with conventional algorithms, leaving unified generative approaches largely unexplored in the literature. Addressing this gap, we propose a unified generative framework to enhance LiDAR simulation fidelity. Our proposed framework projects LiDAR point clouds into depth-reflectance images via a lossless transformation, and employs our novel Controllable Lidar point cloud Generative model, CoLiGen, to translate the images. We extensively evaluate our CoLiGen model, comparing it with the state-of-the-art image-to-image translation models using various metrics to assess the realness, faithfulness, and performance of a downstream perception model. Our results show that CoLiGen exhibits superior performance across most metrics. The dataset and source code for this research are available at https://github.com/hamedhaghighi/CoLiGen.git.
- 19 December 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Sensors Journal
- Vol. 26 (3) , 5013-5025
A Unified Generative Framework for Realistic LiDAR Simulation in Autonomous Driving Systems
Simulation models for perception sensors are integral components of automotive simulators used for the virtual validation of Autonomous Driving Systems (ADS). They also function as powerful tools for generating synthetic datasets to train deep learning-based perception models. LiDAR is a widely adopted perception sensor in ADS due to its high precision in 3D environment scanning. However, developing realistic LiDAR simulation models is a significant technical challenge. Unrealistic simulations can result in a large gap between the synthesised and real-world point clouds, limiting their effectiveness in perception and planning modules. Recently, deep generative models have emerged as promising solutions to synthesise realistic sensory data. However, for LiDAR simulation, deep generative models have been primarily hybridised with conventional algorithms, leaving unified generative approaches largely unexplored in the literature. Addressing this gap, we propose a unified generative framework to enhance LiDAR simulation fidelity. Our proposed framework projects LiDAR point clouds into depth-reflectance images via a lossless transformation, and employs our novel Controllable Lidar point cloud Generative model, CoLiGen, to translate the images. We extensively evaluate our CoLiGen model, comparing it with the state-of-the-art image-to-image translation models using various metrics to assess the realness, faithfulness, and performance of a downstream perception model. Our results show that CoLiGen exhibits superior performance across most metrics. The dataset and source code for this research are available at https://github.com/hamedhaghighi/CoLiGen.git.
- 25 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Artificial Intelligence
- Vol. 7 (6) , 3083-3103
Vetaverse: Technologies, Applications, and Visions toward the Intersection of Metaverse, AI, Vehicles, and Transportation Systems
Since 2021, the term “Metaverse” has been the most popular, garnering much interest. Due to its contained environment and built-in computing and networking capabilities, a modern car makes an intriguing location to host its own little metaverse. Also, travelers don’t have much to do to pass the time while traveling, making them ideal customers for immersive services. Metaverseerse is envisioned as a blended immersive realm that scales to cities and countries, as digital twins of the intelligent Transportation Systems, referred to as “TS-Metaverse”, as well as customized XR services inside each Individual Vehicle, referred to as “IV-Metaverse”. The two subcategories serve fundamentally different purposes, namely longterm interconnection, maintenance, monitoring, and management on scale for large transportation systems (TS), and personalized, private, and immersive infotainment services (IV). By outlining the framework of Vetaverse and examining important AI and facilitating technologies, we reveal this impending trend. Lastly, we examine unresolved issues and potential routes for future study while highlighting some intriguing Vetaverse services.
- 25 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Artificial Intelligence
- Vol. 7 (6) , 3083-3103
Vetaverse: Technologies, Applications, and Visions toward the Intersection of Metaverse, AI, Vehicles, and Transportation Systems
Since 2021, the term “Metaverse” has been the most popular, garnering much interest. Due to its contained environment and built-in computing and networking capabilities, a modern car makes an intriguing location to host its own little metaverse. Also, travelers don’t have much to do to pass the time while traveling, making them ideal customers for immersive services. Metaverseerse is envisioned as a blended immersive realm that scales to cities and countries, as digital twins of the intelligent Transportation Systems, referred to as “TS-Metaverse”, as well as customized XR services inside each Individual Vehicle, referred to as “IV-Metaverse”. The two subcategories serve fundamentally different purposes, namely longterm interconnection, maintenance, monitoring, and management on scale for large transportation systems (TS), and personalized, private, and immersive infotainment services (IV). By outlining the framework of Vetaverse and examining important AI and facilitating technologies, we reveal this impending trend. Lastly, we examine unresolved issues and potential routes for future study while highlighting some intriguing Vetaverse services.
- 6 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 26 (11) , 18283-18303
A Survey and New Perspective of Sensing in the Dark for Intelligent Transportation Systems
Dark or low-luminance environments pose significant challenges for intelligent transportation systems (ITS), potentially affecting perception, decision-making, and safety. In this context, the concept of Vehicular Sensing in the Dark (DarkSensing) is introduced as the use of sensing technologies and algorithms to enable vehicles to perceive, communicate and navigate under dark conditions. It encompasses both individual sensor capabilities and cooperative sensing strategies across all vehicle automation levels. This survey provides the first comprehensive and structured overview of DarkSensing in ITS, focusing on five perception sensor modalities: cameras, LiDAR, RADAR, thermal sensors, and event-based sensors. Specifically, the survey systematically examines the challenges and recent advancements in sensor hardware, perception algorithms, data enhancement methods, and collaborative sensing frameworks. Furthermore, we highlight critical open issues with case studies and outline future research directions. This work establishes the conceptual foundation for DarkSensing and aims to guide researchers and practitioners toward robust ITS performance in dark conditions. Project link: https://github.com/Warwick-Jocelyn/DarkSensing
- 6 October 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Intelligent Transportation Systems
- Vol. 26 (11) , 18283-18303
A Survey and New Perspective of Sensing in the Dark for Intelligent Transportation Systems
Dark or low-luminance environments pose significant challenges for intelligent transportation systems (ITS), potentially affecting perception, decision-making, and safety. In this context, the concept of Vehicular Sensing in the Dark (DarkSensing) is introduced as the use of sensing technologies and algorithms to enable vehicles to perceive, communicate and navigate under dark conditions. It encompasses both individual sensor capabilities and cooperative sensing strategies across all vehicle automation levels. This survey provides the first comprehensive and structured overview of DarkSensing in ITS, focusing on five perception sensor modalities: cameras, LiDAR, RADAR, thermal sensors, and event-based sensors. Specifically, the survey systematically examines the challenges and recent advancements in sensor hardware, perception algorithms, data enhancement methods, and collaborative sensing frameworks. Furthermore, we highlight critical open issues with case studies and outline future research directions. This work establishes the conceptual foundation for DarkSensing and aims to guide researchers and practitioners toward robust ITS performance in dark conditions. Project link: https://github.com/Warwick-Jocelyn/DarkSensing
- 3 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automation Science and Engineering
- Vol. PP (15455955) , 1
Robustness of Panoptic Segmentation for Degraded Automotive Cameras Data
Precise situational awareness is essential for the safe deployment of artificial intelligence in real-world applications, particularly in assisted and automated driving (AAD) systems. Among perception techniques, panoptic segmentation is a promising technique to identify and categorise objects, impending hazards, and drivable space at a pixel level. While panoptic quality might be affected by automotive camera data quality, a comprehensive understanding and modelling of their relationship remains underexplored. Motivated by such a need, this work proposes a unifying pipeline to evaluate the robustness of panoptic segmentation models for automotive cameras, correlating it with 8 traditional image quality metrics (IQA). The proposed pipeline begins by generating a novel degraded dataset, D-Cityscapes+, featuring 19 realistic automotive degradation types at varying severity levels, including novel models for darkness and snowfall conditions with veiling effect. Evaluations on 14 state-of-the-art segmentation model backbones yielded key insights: 1) large-particle degradations (e.g., lens droplets, heavy snow) severely degrade segmentation performance, increasing uncertainty and edge-concentrated segmentation errors; 2) Transformer-based models outperform CNN models under adverse conditions; however, longer processing time, a higher number of parameters, and computational cost are limiting their real-world deployment; 3) frequency-based IQA metrics, such as CW-SSIM, strongly correlate with segmentation performance, serving as reliable predictive tools. 4) visual enhancements via restoration do not coherently benefit downstream segmentation tasks, underscoring the need for perception-specific restoration techniques. The benchmark and code:https://github.com/Warwick-Jocelyn/BRPS.
- 3 September 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Automation Science and Engineering
- Vol. PP (15455955) , 1
Robustness of Panoptic Segmentation for Degraded Automotive Cameras Data
Precise situational awareness is essential for the safe deployment of artificial intelligence in real-world applications, particularly in assisted and automated driving (AAD) systems. Among perception techniques, panoptic segmentation is a promising technique to identify and categorise objects, impending hazards, and drivable space at a pixel level. While panoptic quality might be affected by automotive camera data quality, a comprehensive understanding and modelling of their relationship remains underexplored. Motivated by such a need, this work proposes a unifying pipeline to evaluate the robustness of panoptic segmentation models for automotive cameras, correlating it with 8 traditional image quality metrics (IQA). The proposed pipeline begins by generating a novel degraded dataset, D-Cityscapes+, featuring 19 realistic automotive degradation types at varying severity levels, including novel models for darkness and snowfall conditions with veiling effect. Evaluations on 14 state-of-the-art segmentation model backbones yielded key insights: 1) large-particle degradations (e.g., lens droplets, heavy snow) severely degrade segmentation performance, increasing uncertainty and edge-concentrated segmentation errors; 2) Transformer-based models outperform CNN models under adverse conditions; however, longer processing time, a higher number of parameters, and computational cost are limiting their real-world deployment; 3) frequency-based IQA metrics, such as CW-SSIM, strongly correlate with segmentation performance, serving as reliable predictive tools. 4) visual enhancements via restoration do not coherently benefit downstream segmentation tasks, underscoring the need for perception-specific restoration techniques. The benchmark and code:https://github.com/Warwick-Jocelyn/BRPS.
- Open Access
- 21 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Vehicular Technology
- Vol. 6, 2709-2720
Data-Driven Headlight Flare Model for Automotive Cameras
Automotive cameras are cost effective among perception sensors and have been widely deployed on cars, however the measurements generated by cameras can be affected by various noise factors. Flare, also known as straylight, is a common noise factor especially during night. Automotive headlights can be dazzling for human drivers and might be also challenging for Assisted and Automated Driving (AAD) functions. To enable higher levels of driving automation, investigating and testing this noise factor can be key to achieve AAD in challenging lighting conditions. Therefore, accurate automotive camera flare models need to be thoroughly investigated and developed. However, current camera datasets lack accuracy in representing state-of-the-art automotive cameras. This paper develops, describes, and validates an automotive specific parametrised method for modelling flare induced by automotive headlights. The presented model is validated with real-data and can be fine tuned to fit different types of automotive cameras and headlights. Additionally, this paper introduces a method to seamlessly integrate the modelling results into images generated by simulation platforms, such as CARLA and IPG CarMaker. Using the newly proposed model, automotive datasets with and without the realistic headlight flare can be generated. Overall, the integration of modelled flare provides a framework for accelerating the simulation and testing of assisted and automated driving functions.
- 21 August 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Open Journal of Vehicular Technology
- Vol. 6, 2709-2720
Open AccessData-Driven Headlight Flare Model for Automotive Cameras
Automotive cameras are cost effective among perception sensors and have been widely deployed on cars, however the measurements generated by cameras can be affected by various noise factors. Flare, also known as straylight, is a common noise factor especially during night. Automotive headlights can be dazzling for human drivers and might be also challenging for Assisted and Automated Driving (AAD) functions. To enable higher levels of driving automation, investigating and testing this noise factor can be key to achieve AAD in challenging lighting conditions. Therefore, accurate automotive camera flare models need to be thoroughly investigated and developed. However, current camera datasets lack accuracy in representing state-of-the-art automotive cameras. This paper develops, describes, and validates an automotive specific parametrised method for modelling flare induced by automotive headlights. The presented model is validated with real-data and can be fine tuned to fit different types of automotive cameras and headlights. Additionally, this paper introduces a method to seamlessly integrate the modelling results into images generated by simulation platforms, such as CARLA and IPG CarMaker. Using the newly proposed model, automotive datasets with and without the realistic headlight flare can be generated. Overall, the integration of modelled flare provides a framework for accelerating the simulation and testing of assisted and automated driving functions.
- 27 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Artificial Intelligence
- Vol. 5 (9) , 4765-4775
ClassLIE: Structure- and Illumination-Adaptive Classification for Low-Light Image Enhancement
Low-light images often suffer from limited visibility and multiple types of degradation, rendering low-light image enhancement (LIE) a non-trivial task. Some endeavors have been made to enhance low-light images using convolutional neural networks (CNNs). However, they have low efficiency in learning the structural information and diverse illumination levels at the local regions of an image. Consequently, the enhanced results are affected by unexpected artifacts, such as unbalanced exposure, blur, and color bias. This paper proposes a novel framework, called ClassLIE, that combines the potential of CNNs and transformers. It classifies and adaptively learns the structural and illumination information from the low-light images in a holistic and regional manner, thus showing better enhancement performance. Our framework first employs a structure and illumination classification (SIC) module to learn the degradation information adaptively. In SIC, we decompose an input image into an illumination map and a reflectance map. A class prediction block is then designed to classify the degradation information by calculating the structure similarity scores on the reflectance map and mean square error on the illumination map. As such, each input image can be divided into patches with three enhancement difficulty levels. Then, a feature learning and fusion (FLF) module is proposed to adaptively learn the feature information with CNNs for different enhancement difficulty levels while learning the long-range dependencies for the patches in a holistic manner. Experiments on five benchmark datasets consistently show our ClassLIE achieves new state-of-the-art performance, with 25.74 PSNR and 0.92 SSIM on the LOL dataset.
- 27 May 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Artificial Intelligence
- Vol. 5 (9) , 4765-4775
ClassLIE: Structure- and Illumination-Adaptive Classification for Low-Light Image Enhancement
Low-light images often suffer from limited visibility and multiple types of degradation, rendering low-light image enhancement (LIE) a non-trivial task. Some endeavors have been made to enhance low-light images using convolutional neural networks (CNNs). However, they have low efficiency in learning the structural information and diverse illumination levels at the local regions of an image. Consequently, the enhanced results are affected by unexpected artifacts, such as unbalanced exposure, blur, and color bias. This paper proposes a novel framework, called ClassLIE, that combines the potential of CNNs and transformers. It classifies and adaptively learns the structural and illumination information from the low-light images in a holistic and regional manner, thus showing better enhancement performance. Our framework first employs a structure and illumination classification (SIC) module to learn the degradation information adaptively. In SIC, we decompose an input image into an illumination map and a reflectance map. A class prediction block is then designed to classify the degradation information by calculating the structure similarity scores on the reflectance map and mean square error on the illumination map. As such, each input image can be divided into patches with three enhancement difficulty levels. Then, a feature learning and fusion (FLF) module is proposed to adaptively learn the feature information with CNNs for different enhancement difficulty levels while learning the long-range dependencies for the patches in a holistic manner. Experiments on five benchmark datasets consistently show our ClassLIE achieves new state-of-the-art performance, with 25.74 PSNR and 0.92 SSIM on the LOL dataset.
- Open Access
Exploring Generative AI for Sim2Real in Driving Data Synthesis
Datasets are essential for training and testing vehicle perception algorithms. However, the collection and annotation of real-world images is time-consuming and expensive. Driving simulators offer a solution by automatically generating various driving scenarios with corresponding annotations, but the simulation-to-reality (Sim2Real) domain gap remains a challenge. While most of the Generative Artificial Intelligence (AI) follows the de facto Generative Adversarial Nets (GANs)-based methods, the recent emerging diffusion probabilistic models have not been fully explored in mitigating Sim2Real challenges for driving data synthesis. To explore the performance, this paper applied three different generative AI methods to leverage semantic label maps from a driving simulator as a bridge for the creation of realistic datasets. A comparative analysis of these methods is presented from the perspective of image quality and perception. New synthetic datasets, which include driving images and auto-generated high-quality annotations, are produced with low costs and high scene variability. The experimental results show that although GAN-based methods are adept at generating high-quality images when provided with manually annotated labels, ControlNet produces synthetic datasets with fewer artefacts and more structural fidelity when using simulator-generated labels. This suggests that the diffusion-based approach may provide improved stability and an alternative method for addressing Sim2Real challenges.
Open AccessExploring Generative AI for Sim2Real in Driving Data Synthesis
Datasets are essential for training and testing vehicle perception algorithms. However, the collection and annotation of real-world images is time-consuming and expensive. Driving simulators offer a solution by automatically generating various driving scenarios with corresponding annotations, but the simulation-to-reality (Sim2Real) domain gap remains a challenge. While most of the Generative Artificial Intelligence (AI) follows the de facto Generative Adversarial Nets (GANs)-based methods, the recent emerging diffusion probabilistic models have not been fully explored in mitigating Sim2Real challenges for driving data synthesis. To explore the performance, this paper applied three different generative AI methods to leverage semantic label maps from a driving simulator as a bridge for the creation of realistic datasets. A comparative analysis of these methods is presented from the perspective of image quality and perception. New synthetic datasets, which include driving images and auto-generated high-quality annotations, are produced with low costs and high scene variability. The experimental results show that although GAN-based methods are adept at generating high-quality images when provided with manually annotated labels, ControlNet produces synthetic datasets with fewer artefacts and more structural fidelity when using simulator-generated labels. This suggests that the diffusion-based approach may provide improved stability and an alternative method for addressing Sim2Real challenges.
- Open Access
Benchmarking the Robustness of Panoptic Segmentation for Automated Driving
Precise situational awareness is required for the safe decision-making of assisted and automated driving (AAD) functions. Panoptic segmentation is a promising perception technique to identify and categorise objects, impending hazards, and driveable space at a pixel level. While segmentation quality is generally associated with the quality of the camera data, a comprehensive understanding and modelling of this relationship are paramount for AAD system designers. Motivated by such a need, this work proposes a unifying pipeline to assess the robustness of panoptic segmentation models for AAD, correlating it with traditional image quality. The first step of the proposed pipeline involves generating degraded camera data that reflects real-world noise factors. To this end, 19 noise factors have been identified and implemented with 3 severity levels. Of these factors, this work proposes novel models for unfavourable light and snow. After applying the degradation models, three state-of-the-art CNN- and vision transformers (ViT)-based panoptic segmentation networks are used to analyse their robustness. The variations of the segmentation performance are then correlated to 8 selected image quality metrics. This research reveals that: 1) certain specific noise factors produce the highest impact on panoptic segmentation, i.e. droplets on lens and Gaussian noise; 2) the ViT-based panoptic segmentation backbones show better robustness to the considered noise factors; 3) some image quality metrics (i.e. LPIPS and CW-SSIM) correlate strongly with panoptic segmentation performance and therefore they can be used as predictive metrics for network performance.
Open AccessBenchmarking the Robustness of Panoptic Segmentation for Automated Driving
Precise situational awareness is required for the safe decision-making of assisted and automated driving (AAD) functions. Panoptic segmentation is a promising perception technique to identify and categorise objects, impending hazards, and driveable space at a pixel level. While segmentation quality is generally associated with the quality of the camera data, a comprehensive understanding and modelling of this relationship are paramount for AAD system designers. Motivated by such a need, this work proposes a unifying pipeline to assess the robustness of panoptic segmentation models for AAD, correlating it with traditional image quality. The first step of the proposed pipeline involves generating degraded camera data that reflects real-world noise factors. To this end, 19 noise factors have been identified and implemented with 3 severity levels. Of these factors, this work proposes novel models for unfavourable light and snow. After applying the degradation models, three state-of-the-art CNN- and vision transformers (ViT)-based panoptic segmentation networks are used to analyse their robustness. The variations of the segmentation performance are then correlated to 8 selected image quality metrics. This research reveals that: 1) certain specific noise factors produce the highest impact on panoptic segmentation, i.e. droplets on lens and Gaussian noise; 2) the ViT-based panoptic segmentation backbones show better robustness to the considered noise factors; 3) some image quality metrics (i.e. LPIPS and CW-SSIM) correlate strongly with panoptic segmentation performance and therefore they can be used as predictive metrics for network performance.