Mengqi Ji
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Mengqi Ji has a total of 49 co-authors. Here are the top co-authors by number of shared publications:
- Lu Fang
Department of Electronic Engineering, Beijing National Research Center for Information Science and Technology, Tsinghua University, Beijing, China
5 shared publications Feng Duan
Interventional Radiology Department, The First Medicine Centre, Chinese PLA General Hospital, Beijing, China
3 shared publicationsJingtao Fan
Department of Automation, Tsinghua University, Beijing, China
3 shared publicationsLaiqiang Huang
Tsinghua–Berkeley Shenzhen Institute (TBSI), Tsinghua Shenzhen International Graduate School (SIGS), Tsinghua University, Shenzhen, China
3 shared publications- Qionghai Dai
Department of Automation, Tsinghua University, Beijing, China
3 shared publications Shengwei Jiang
Tsinghua–Berkeley Shenzhen Institute (TBSI), Tsinghua Shenzhen International Graduate School (SIGS), Tsinghua University, Shenzhen, China
3 shared publications
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 5252-5260
SplineHMR: A Universal B-spline Prior for Temporally Stable SMPL Refinement
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 5252-5260
Open AccessSplineHMR: A Universal B-spline Prior for Temporally Stable SMPL Refinement
- Open Access
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 5686-5697
DREAM: LLM-based Dynamic Role-playing via Event-Aware Memory Graph
- 8 August 2026
- Published by Association for Computing Machinery (ACM)
- p. 5686-5697
Open AccessDREAM: LLM-based Dynamic Role-playing via Event-Aware Memory Graph
- Open Access
- 3 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 76-82
Phase-Aligned Template and Residual-Guided Unsupervised Anomaly Detection for Aero-Engine Blade Borescope Inspection
- 3 April 2026
- Published by Association for Computing Machinery (ACM)
- p. 76-82
Open AccessPhase-Aligned Template and Residual-Guided Unsupervised Anomaly Detection for Aero-Engine Blade Borescope Inspection
- Open Access
- 18 September 2024
- Published by Institution of Engineering and Technology (IET) in CAAI Transactions on Intelligence Technology
- Vol. 10 (1) , 193-206
Visible and near‐infrared image fusion based on information complementarity
Images with complementary spectral information can be recorded using image sensors that can identify visible and near‐infrared spectrum. The fusion of visible and near‐infrared (NIR) aims to enhance the quality of images acquired by video monitoring systems for the ease of user observation and data processing. Unfortunately, current fusion algorithms produce artefacts and colour distortion since they cannot make use of spectrum properties and are lacking in information complementarity. Therefore, an information complementarity fusion (ICF) model is designed based on physical signals. In order to separate high‐frequency noise from important information in distinct frequency layers, the authors first extracted texture‐scale and edge‐scale layers using a two‐scale filter. Second, the difference map between visible and near‐infrared was filtered using the extended‐DoG filter to produce the initial visible‐NIR complementary weight map. Then, to generate a guide map, the near‐infrared image with night adjustment was processed as well. The final complementarity weight map was subsequently derived via an arctanI function mapping using the guide map and the initial weight maps. Finally, fusion images were generated with the complementarity weight maps. The experimental results demonstrate that the proposed approach outperforms the state‐of‐the‐art in both avoiding artificial colours as well as effectively utilising information complementarity.
- 18 September 2024
- Published by Institution of Engineering and Technology (IET) in CAAI Transactions on Intelligence Technology
- Vol. 10 (1) , 193-206
Open AccessVisible and near‐infrared image fusion based on information complementarity
Images with complementary spectral information can be recorded using image sensors that can identify visible and near‐infrared spectrum. The fusion of visible and near‐infrared (NIR) aims to enhance the quality of images acquired by video monitoring systems for the ease of user observation and data processing. Unfortunately, current fusion algorithms produce artefacts and colour distortion since they cannot make use of spectrum properties and are lacking in information complementarity. Therefore, an information complementarity fusion (ICF) model is designed based on physical signals. In order to separate high‐frequency noise from important information in distinct frequency layers, the authors first extracted texture‐scale and edge‐scale layers using a two‐scale filter. Second, the difference map between visible and near‐infrared was filtered using the extended‐DoG filter to produce the initial visible‐NIR complementary weight map. Then, to generate a guide map, the near‐infrared image with night adjustment was processed as well. The final complementarity weight map was subsequently derived via an arctanI function mapping using the guide map and the initial weight maps. Finally, fusion images were generated with the complementarity weight maps. The experimental results demonstrate that the proposed approach outperforms the state‐of‐the‐art in both avoiding artificial colours as well as effectively utilising information complementarity.
- Open Access
- 23 August 2023
- Published by Elsevier in Engineering
- Vol. 34, 70-82
The Group Interaction Field for Learning and Explaining Pedestrian Anticipation
- 23 August 2023
- Published by Elsevier in Engineering
- Vol. 34, 70-82
Open AccessThe Group Interaction Field for Learning and Explaining Pedestrian Anticipation
- 23 February 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 30 (10577149) , 3240-3251
Boosting Single Image Super-Resolution Learnt From Implicit Multi-Image Prior
Learning-based single image super-resolution (SISR) aims to learn a versatile mapping from low resolution (LR) image to its high resolution (HR) version. The critical challenge is to bias the network training towards continuous and sharp edges. For the first time in this work, we propose an implicit boundary prior learnt from multi-view observations to significantly mitigate the challenge in SISR we outline. Specifically, the multi-image prior that encodes both disparity information and boundary structure of the scene supervise a SISR network for edge-preserving. For simplicity, in the training procedure of our framework, light field (LF) serves as an effective multi-image prior, and a hybrid loss function jointly considers the content, structure, variance as well as disparity information from 4D LF data. Consequently, for inference, such a general training scheme boosts the performance of various SISR networks, especially for the regions along edges. Extensive experiments on representative backbone SISR architectures constantly show the effectiveness of the proposed method, leading to around 0.6 dB gain without modifying the network architecture.
- 23 February 2021
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Image Processing
- Vol. 30 (10577149) , 3240-3251
Boosting Single Image Super-Resolution Learnt From Implicit Multi-Image Prior
Learning-based single image super-resolution (SISR) aims to learn a versatile mapping from low resolution (LR) image to its high resolution (HR) version. The critical challenge is to bias the network training towards continuous and sharp edges. For the first time in this work, we propose an implicit boundary prior learnt from multi-view observations to significantly mitigate the challenge in SISR we outline. Specifically, the multi-image prior that encodes both disparity information and boundary structure of the scene supervise a SISR network for edge-preserving. For simplicity, in the training procedure of our framework, light field (LF) serves as an effective multi-image prior, and a hybrid loss function jointly considers the content, structure, variance as well as disparity information from 4D LF data. Consequently, for inference, such a general training scheme boosts the performance of various SISR networks, especially for the regions along edges. Extensive experiments on representative backbone SISR architectures constantly show the effectiveness of the proposed method, leading to around 0.6 dB gain without modifying the network architecture.
- 8 January 2021
- Published by Springer Nature in Nature Machine Intelligence
- Vol. 3 (1) , 97
Author Correction: Augmenting vascular disease diagnosis by vasculature-aware unsupervised learning
- 8 January 2021
- Published by Springer Nature in Nature Machine Intelligence
- Vol. 3 (1) , 97
Author Correction: Augmenting vascular disease diagnosis by vasculature-aware unsupervised learning
- 1 June 2020
- Published by Springer Nature in Nature Machine Intelligence
- Vol. 2 (6) , 337-346
Augmenting vascular disease diagnosis by vasculature-aware unsupervised learning
- 1 June 2020
- Published by Springer Nature in Nature Machine Intelligence
- Vol. 2 (6) , 337-346
Augmenting vascular disease diagnosis by vasculature-aware unsupervised learning
- 25 May 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 43 (11) , 4078-4093
SurfaceNet+: An End-to-end 3D Neural Network for Very Sparse Multi-View Stereopsis
Multi-view stereopsis (MVS) tries to recover the 3D model from 2D images. As the observations become sparser, the significant 3D information loss makes the MVS problem more challenging. Instead of only focusing on densely sampled conditions, we investigate sparse-MVS with large baseline angles since the sparser sensation is more practical and more cost-efficient. By investigating various observation sparsities, we show that the classical depth-fusion pipeline becomes powerless for the case with a larger baseline angle that worsens the photo-consistency check. As another line of the solution, we present SurfaceNet+, a volumetric method to handle the ‘incompleteness’ and the ‘inaccuracy’ problems induced by a very sparse MVS setup. Specifically, the former problem is handled by a novel volume-wise view selection approach. It owns superiority in selecting valid views while discarding invalid occluded views by considering the geometric prior. Furthermore, the latter problem is handled via a multi-scale strategy that consequently refines the recovered geometry around the region with the repeating pattern. The experiments demonstrate the tremendous performance gap between SurfaceNet+ and state-of-the-art methods in terms of precision and recall. Under the extreme sparse-MVS settings in two datasets, where existing methods can only return very few points, SurfaceNet+ still works as well as in the dense MVS setting.
- 25 May 2020
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Pattern Analysis and Machine Intelligence
- Vol. 43 (11) , 4078-4093
SurfaceNet+: An End-to-end 3D Neural Network for Very Sparse Multi-View Stereopsis
Multi-view stereopsis (MVS) tries to recover the 3D model from 2D images. As the observations become sparser, the significant 3D information loss makes the MVS problem more challenging. Instead of only focusing on densely sampled conditions, we investigate sparse-MVS with large baseline angles since the sparser sensation is more practical and more cost-efficient. By investigating various observation sparsities, we show that the classical depth-fusion pipeline becomes powerless for the case with a larger baseline angle that worsens the photo-consistency check. As another line of the solution, we present SurfaceNet+, a volumetric method to handle the ‘incompleteness’ and the ‘inaccuracy’ problems induced by a very sparse MVS setup. Specifically, the former problem is handled by a novel volume-wise view selection approach. It owns superiority in selecting valid views while discarding invalid occluded views by considering the geometric prior. Furthermore, the latter problem is handled via a multi-scale strategy that consequently refines the recovered geometry around the region with the repeating pattern. The experiments demonstrate the tremendous performance gap between SurfaceNet+ and state-of-the-art methods in terms of precision and recall. Under the extreme sparse-MVS settings in two datasets, where existing methods can only return very few points, SurfaceNet+ still works as well as in the dense MVS setting.
- 1 March 2020
- Published by The Hong Kong University of Science and Technology Library
Learning to perceive 3D surface
- 1 March 2020
- Published by The Hong Kong University of Science and Technology Library
Learning to perceive 3D surface