Dr Li Sun
PhD
Department of Computer Science
Visiting academic
Member of the Visual Computing research group
+44 114 222 1935
Full contact details
Department of Computer Science
Regent Court (DCS)
211 Portobello
Sheffield
S1 4DP
- Profile
-
Li Sun (Kevin) was a Lecturer of Computer Science at University of Sheffield, since September 2021 he holds a visiting academic position. He was a Postdoctoral Research Fellow at the Oxford Robotics Institute, University of Oxford.
Kevin was awarded the prestigious national scholarship from the Scottish government and received PhD in robotics from the University of Glasgow in 2016. He worked as a postdoctoral research fellow at Lincoln Centre of Autonomous Systems, University of Lincoln (2017-2018) and Extreme Robotics Lab, University of Birmingham (2016-2017).
Since 2012, he has worked as a main researcher/contributor of several EU/EPSRC funded projects e.g. CloPeMa (€3.7 m), RoMaNs (€6.4 m), ILIAD (€6.9 m), and MobileRobotics (£4.9 m).
- Research interests
-
Kevin's research focuses on the core challenges in the emerging robot vision to enable the robot to manipulate complex industrial objects eg deformable object and waste-like objects or drive in the dynamic, real-life environments eg warehouses, on-road/off-road driving. His research pursues the fantastic intersection between statistical machine learning and solid robotic system integration.
Research interests
- Robotics and Autonomous Systems, Computer vision and machine learning.
Application fields
- multi-sensors perception i.e. detection, recognition, tracking using RGB-D, LiDAR sensors
- 3D mapping and long-term robot localisation in changing environments
- autonomous visually-guided robot manipulation and navigation
- Publications
-
Journal articles
- Long short‐distance topology modelling of 3D point cloud segmentation with a graph convolution neural network. IET Computer Vision.
- Deep 3D caricature face generation with identity and structure consistency. Neurocomputing.
- Design, Analysis and Implementation of a Smart Next Generation Secure Shipping Infrastructure Using Autonomous Robot. Computer Networks, 107779-107779.
- A survey on underactuated robotic systems: Bio-inspiration, trajectory planning and control. Mechatronics, 72, 102443-102443.
- Local k-NNs pattern in omni-direction graph convolution neural network for 3D point clouds. Neurocomputing, 413, 487-498.
- Simultaneous material segmentation and 3D reconstruction in industrial scenarios. Frontiers in Robotics and AI, 7. View this article in WRRO
- A self-propelled robotic system with a visco-elastic joint: dynamics and motion analysis. Engineering with Computers, 36(2), 655-669.
- Generative localisation with uncertainty estimation through video-CT data for bronchoscopic biopsy. IEEE Robotics and Automation Letters. View this article in WRRO
- Learning Kalman network : a deep monocular visual odometry for on-road driving. Robotics and Autonomous Systems, 121. View this article in WRRO
- Continuous perception for deformable objects understanding. Robotics and Autonomous Systems, 118, 220-230. View this article in WRRO
- A novel weakly-supervised approach for RGB-D-based nuclear waste object detection. IEEE Sensors Journal, 19(9), 3487-3500. View this article in WRRO
- Autonomous clothes manipulation using a hierarchical vision architecture. IEEE Access, 6, 76646-76662. View this article in WRRO
- Recurrent-OctoMap: Learning state-based map refinement for long-term semantic mapping with 3-D-Lidar data. IEEE Robotics and Automation Letters, 3(4), 3749-3756. View this article in WRRO
- Integrating Deep Semantic Segmentation Into 3-D Point Cloud Registration. IEEE Robotics and Automation Letters, 3(4), 2942-2949.
- A Precise Method for Cloth Configuration Parsing Applied to Single-Arm Flattening. International Journal of Advanced Robotic Systems, 13(2), 70-70.
- ECG Analysis Using Multiple Instance Learning for Myocardial Infarction Detection. IEEE Transactions on Biomedical Engineering, 59(12), 3348-3356.
- Dense RGB-D Semantic Mapping with Pixel-Voxel Neural Network. Sensors, 18(9), 3099-3099.
- Introspective Visuomotor Control: Exploiting Uncertainty in Deep Visuomotor Control for Failure Recovery.
Conference proceedings papers
- EU Long-term Dataset with Multiple Sensors for Autonomous Driving. Proceedings of the ... IEEE/RSJ International Conference on Intelligent Robots and Systems. IEEE/RSJ International Conference on Intelligent Robots and Systems, Vol. 00 (pp 10697-10704)
- Learning Monocular Visual Odometry with Dense 3D Mapping from Dense 3D Flow. 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 1 October 2018 - 5 October 2018.
- Multisensor Online Transfer Learning for 3D LiDAR-Based Human Detection with a Mobile Robot. 2018 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 1 October 2018 - 5 October 2018.
- 3DOF Pedestrian Trajectory Prediction Learned from Long-Term Autonomous Mobile Robot Deployment Data. 2018 IEEE International Conference on Robotics and Automation (ICRA), 21 May 2018 - 25 May 2018.
- A fully end-to-end deep learning approach for real-time simultaneous 3D reconstruction and material recognition. 2017 18th International Conference on Advanced Robotics (ICAR), 10 July 2017 - 12 July 2017.
- Single-shot clothing category recognition in free-configurations with application to autonomous clothes sorting. 2017 IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS), 24 September 2017 - 28 September 2017.
- Interactive perception based on Gaussian Process classification for house-hold objects recognition & sorting. 2016 IEEE International Conference on Robotics and Biomimetics (ROBIO), 3 December 2016 - 7 December 2016.
- Recognising the clothing categories from free-configuration using Gaussian-Process-based interactive perception. 2016 IEEE International Conference on Robotics and Automation (ICRA), 16 May 2016 - 21 May 2016.
- Accurate garment surface analysis using an active stereo robot head with application to dual-arm flattening. 2015 IEEE International Conference on Robotics and Automation (ICRA), 26 May 2015 - 30 May 2015.
- Localising Faster: Efficient and precise lidar-based robot localisation in large-scale environments. IEEE International Conference on Robotics and Automation : ICRA : [proceedings] IEEE International Conference on Robotics and Automation View this article in WRRO
- Introspective Visuomotor Control: Exploiting Uncertainty in Deep Visuomotor Control for Failure Recovery. IEEE International Conference on Robotics and Automation : ICRA : [proceedings] IEEE International Conference on Robotics and Automation
- A Quantifiable Stratification Method for Tidy-up in Service Robotics. IEEE
Preprints
- View this article in WRRO Imagine That! Leveraging Emergent Affordances for 3D Tool Synthesis.
- EPANer Team Description Paper for World Robot Challenge 2020.
- End-to-end Recurrent Multi-Object Tracking and Trajectory Prediction with Relational Reasoning.
- Robot Vision Architecture for Autonomous Clothes Manipulation.
- Weakly-supervised DCNN for RGB-D Object Recognition in Real-World Applications Which Lack Large-scale Annotated Training Data.
- Zero-shot Anomalous Object Detection using Unsupervised Metric Learning.
- View this article in WRRO NDT-Transformer: Large-Scale 3D Point Cloud Localisation using the Normal Distribution Transform Representation.
- Grants
-
Current Grants
- Towards long-term mapping and semantic guided robot navigation in changing environments, Royal Society, 10/2020 - 09/2021, £19,934, as PI
- Advanced 3D mapping and long-term localisation with semantics understanding and terrain assessment for the planetary robot navigation, UKRI, 03/2020 - 09/2021, £144,950, as PI
- Professional activities and memberships
-
Kevin is a member of IEEE, BMVA, EUcog, SICSA. He serves as reviewers in top tier robotics and AI conferences e.g. RSS, ICRA, IROS, NIPS, AAAI and journals eg IJRR, TRO, RA-L. He is a member of the Visual Computing research group for the University of Sheffield.