Yuhao Sun
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
Yuhao Sun has a total of 18 co-authors. Here are the top co-authors by number of shared publications:
- Bin Fang
Beijing University of Posts and Telecommunications
3 shared publications - Fuchun Sun
Department of Computer Science and Technology, Tsinghua University, Beijing 100084, China
2 shared publications - Shixin Zhang
School of Artificial Intelligence, Beijing University of Posts and Telecommunications, Beijing, China
2 shared publications - Zixi Chen
Vrije Universiteit Brussel (VUB)
2 shared publications - Derui Li
Beijing University of Post and Telecommunications, Beijing, China
1 shared publication Heyi Sun
Zhili College, Tsinghua University, Beijing, China
1 shared publication
Latest Publications
- Open Access
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 4426-4435
Bridging Language and Spherical Space: Object-Centric Control for Text-to-Panorama Generation
- 9 November 2026
- Published by Association for Computing Machinery (ACM)
- p. 4426-4435
Open AccessBridging Language and Spherical Space: Object-Centric Control for Text-to-Panorama Generation
- Open Access
- 31 May 2026
- Published by World Scientific Pub Co Pte Ltd in Guidance, Navigation and Control
- Vol. 06 (02) , 243-246
DeepTouch: Embodied Intelligence with Vision-Tactile Fusion for Dexterous Manipulation
In the field of embodied intelligence, the effective integration of vision and touch is essential for achieving dexterous manipulation. To address the limitations of vision-only perception in perceiving contact states and material properties, we propose DeepTouch, a vision-tactile fusion framework integrating visuo-tactile perception, simulated and real data acquisition, and a vision-tactile-language-action (VTLA) control strategy. It provides a concise technical framework for refined dexterous manipulation.
- 31 May 2026
- Published by World Scientific Pub Co Pte Ltd in Guidance, Navigation and Control
- Vol. 06 (02) , 243-246
Open AccessDeepTouch: Embodied Intelligence with Vision-Tactile Fusion for Dexterous Manipulation
In the field of embodied intelligence, the effective integration of vision and touch is essential for achieving dexterous manipulation. To address the limitations of vision-only perception in perceiving contact states and material properties, we propose DeepTouch, a vision-tactile fusion framework integrating visuo-tactile perception, simulated and real data acquisition, and a vision-tactile-language-action (VTLA) control strategy. It provides a concise technical framework for refined dexterous manipulation.
- 16 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 9 (12) , 11218-11225
Simulation of Optical Tactile Sensors Supporting Slip and Rotation Using Path Tracing and IMPM
Optical tactile sensors are extensively utilized in intelligent robot manipulation due to their ability to acquire high-resolution tactile information at a lower cost. However, achieving adequate reality and versatility in simulating optical tactile sensors is challenging. In this letter, we propose a simulation method and validate its effectiveness through experiments. We utilize path tracing for image rendering, achieving higher similarity to real data than the baseline method in simulating pressing scenarios. Additionally, we apply the improved Material Point Method(IMPM) algorithm to simulate the relative rest between the object and the elastomer surface when the object is in motion, enabling more accurate simulation of complex manipulations such as slip and rotation.
- 16 October 2024
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 9 (12) , 11218-11225
Simulation of Optical Tactile Sensors Supporting Slip and Rotation Using Path Tracing and IMPM
Optical tactile sensors are extensively utilized in intelligent robot manipulation due to their ability to acquire high-resolution tactile information at a lower cost. However, achieving adequate reality and versatility in simulating optical tactile sensors is challenging. In this letter, we propose a simulation method and validate its effectiveness through experiments. We utilize path tracing for image rendering, achieving higher similarity to real data than the baseline method in simulating pressing scenarios. Additionally, we apply the improved Material Point Method(IMPM) algorithm to simulate the relative rest between the object and the elastomer surface when the object is in motion, enabling more accurate simulation of complex manipulations such as slip and rotation.
- 18 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 8 (10) , 6307-6314
TIRgel: A Visuo-Tactile Sensor With Total Internal Reflection Mechanism for External Observation and Contact Detection
This letter proposes a vision-based tactile sensor named TIRgel, leveraging visual integration to simplify the sensing system. First, the sensor achieves conversion of visual and tactile modality via focus adjustment. Under far-focus imaging, the camera can observe external environments; Under near-focus imaging, we perform total internal reflection (TIR) within the lenticular elastomer to visualize tactile features. The effectiveness of the sensing mechanism is verified via MPM (Material Point Method) simulation. Second, the TIRgel can capture and extract micron-scale tactile details based on luminosity information. Tactile imaging has the natural fusion of visual and tactile information, where the luminosity boundary reflects the texture feature, and the background contains the object's appearance information. Inherent data fusion simplifies multimodal data acquisition and reduces the information processing difficulty. Finally, the TIRgel can resist the interference of ambient brightness via internal illumination and has a higher adaptive capacity to the environment than unidirectional perspective VBTSs. We believe that it has the potential to provide visual-tactile integration for robot operation.
- 18 August 2023
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Robotics and Automation Letters
- Vol. 8 (10) , 6307-6314
TIRgel: A Visuo-Tactile Sensor With Total Internal Reflection Mechanism for External Observation and Contact Detection
This letter proposes a vision-based tactile sensor named TIRgel, leveraging visual integration to simplify the sensing system. First, the sensor achieves conversion of visual and tactile modality via focus adjustment. Under far-focus imaging, the camera can observe external environments; Under near-focus imaging, we perform total internal reflection (TIR) within the lenticular elastomer to visualize tactile features. The effectiveness of the sensing mechanism is verified via MPM (Material Point Method) simulation. Second, the TIRgel can capture and extract micron-scale tactile details based on luminosity information. Tactile imaging has the natural fusion of visual and tactile information, where the luminosity boundary reflects the texture feature, and the background contains the object's appearance information. Inherent data fusion simplifies multimodal data acquisition and reduces the information processing difficulty. Finally, the TIRgel can resist the interference of ambient brightness via internal illumination and has a higher adaptive capacity to the environment than unidirectional perspective VBTSs. We believe that it has the potential to provide visual-tactile integration for robot operation.
- Open Access
- 27 September 2021
- Published by Association for Research in Vision and Ophthalmology (ARVO) in Journal of Vision
- Vol. 21 (9) , 2555
High familiar faces have both part recognition and holistic processing advantages
Previous studies reported that people recognize familiar faces better than unfamiliar faces(Jenkins, White, Van Montford & Burton, 2011). Does face part-based and/or holistic processing play a role in it? Wang and her colleagues (2015, 2019) in studies for face other-race effect (ORE) and species-specific effect (SSE) found that both part-based and holistic processing were strengthened, in a region-selective (i.e., eyes region vs. mouth region ) way, by people daily experience accumulation. In this current study, we examined whether both part-based (probably region-selective) and holistic processing contribute to face familiarity effect. Three levels of familiar faces were used. They were classmates’ faces (high familiar), schoolmates in the same department but different classes (middle familiar), and schoolmates in different departments (unfamiliar). Using an old/new task, Experiment 1 found high familiar eyes recognized better than middle familiar or unfamiliar faces (p=0.004, 0.001, respectively), but not for mouth recognition, suggesting a part-based region-selective familiarity effect. Using the “Perceptual field” Paradigm (Van Belle et al., 2015), Experiment 2 found that high familiar faces have strongest inversion effect (ps<0.001, 0.35, in high familiar, middle familiar, and unfamiliar faces conditions, respectively), suggesting face holistic processing plays a role in face familiarity effect. The experience-based holistic processing hypothesis for face recognition was supported.
- 27 September 2021
- Published by Association for Research in Vision and Ophthalmology (ARVO) in Journal of Vision
- Vol. 21 (9) , 2555
Open AccessHigh familiar faces have both part recognition and holistic processing advantages
Previous studies reported that people recognize familiar faces better than unfamiliar faces(Jenkins, White, Van Montford & Burton, 2011). Does face part-based and/or holistic processing play a role in it? Wang and her colleagues (2015, 2019) in studies for face other-race effect (ORE) and species-specific effect (SSE) found that both part-based and holistic processing were strengthened, in a region-selective (i.e., eyes region vs. mouth region ) way, by people daily experience accumulation. In this current study, we examined whether both part-based (probably region-selective) and holistic processing contribute to face familiarity effect. Three levels of familiar faces were used. They were classmates’ faces (high familiar), schoolmates in the same department but different classes (middle familiar), and schoolmates in different departments (unfamiliar). Using an old/new task, Experiment 1 found high familiar eyes recognized better than middle familiar or unfamiliar faces (p=0.004, 0.001, respectively), but not for mouth recognition, suggesting a part-based region-selective familiarity effect. Using the “Perceptual field” Paradigm (Van Belle et al., 2015), Experiment 2 found that high familiar faces have strongest inversion effect (ps<0.001, 0.35, in high familiar, middle familiar, and unfamiliar faces conditions, respectively), suggesting face holistic processing plays a role in face familiarity effect. The experience-based holistic processing hypothesis for face recognition was supported.