We conduct research in the field of Computer Vision, Robotics, and Artificial Intelligence. Our primary interests include:

• Computer Vision: Visual representation learning, visual automation, single-view and multi-view 3D acquisition, camera modeling, light modeling, and image enhancement.
• Robotics: Advancing robotic perception, control, and planning, enabling robots to operate autonomously in complex and unstructured environments.
• Artificial Intelligence: Efficient algorithms for supervised, unsupervised, and reinforcement learning, with a focus on exploring generative AI approaches for generalization, robustness, and scalability.

PI offers following courses at Texas A&M University, College Station Campus

• Foundations of Visual Computing
  Course Website. Fall 2024, 2025
• Generative AI for Art and Content Creation
  Course Website. Spring 2024, 2025, 2026
• Mathematics for Visual Computing
  Course Website. Fall 2026
• Introduction to Research in VCIM
  Course Website. Fall 2026
• Visual Computing and Machine Intelligence
  Course Website. Fall 2027
LAB MEMBERS

---

Faculty / Principal Investigator: Suryansh Kumar


Ph.D. Students: Under search

M.S. Students: Miqdad Naduthodi, Praroop Chanda,
Fred Cui

ALUMNI

[first position following supervision noted]

---

• Rethinking Dense Optical Flow without Test-Time Scaling
  Praroop Chanda, Suryansh Kumar
  Conference on Computer Vision and Pattern Recognition (CVPR), 2026, ViSCALE
  [ paper ] [ code ]
• A Dataset and Evaluation for Complex 4D Markerless Human Motion Capture
  Yeeun Park, Miqdad Naduthodi, Suryansh Kumar
  Conference on Computer Vision and Pattern Recognition (CVPR), 2026, 4D-WM
  [ paper ] [ code ]
• 3DGS for Efficient Retrospective Dynamic Scene Novel View Synthesis with a Standardized Benchmark
  Yunxiao Zhang, Suryansh Kumar
  Conference on Computer Vision and Pattern Recognition (CVPR), 2026, 4D-WM
  [ paper ] [ code ]
• Test-Time Domain Adaptation for Interactive Video Generation
  Ishaan Singh Rawal, Suryansh Kumar
  Conference on Computer Vision and Pattern Recognition (CVPR), 2026, VGBE
  [ paper ] [ code ]
• Time-Archival Camera Virtualization for Sports and Visual Performances
  Yunxiao Zhang, William Stone, Suryansh Kumar
  Computer Vision and Image Understanding (CVIU) 2026
  [ paper ] [ code ]
• Mobile Robotic Multi-View Photometric Stereo
  Suryansh Kumar
  International Journal of Photogrammetry and Remote Sensing (ISPRS) 2025
  [ paper ] [ code ]
• Stereo Risk: A Continuous Modeling Approach to Stereo Matching
  Ce Liu*, Suryansh Kumar*, Shuhang Gu, Radu Timofte, Yao Yao, Luc Van Gool
  International Conference on Machine Leaning (ICML) 2024 [Oral Paper]
  [ paper ] [ code ]
• ICGNet: A Unified Approach for Instance-Centric Grasping
  R. Zurbrugg, Y. Liu, F. Engelmann, Suryansh Kumar, Marco Hutter, V. Patil, F. Yu
  International Conference on Robotics and Automation (ICRA) 2024
  [ paper ] [ code ]
• Learning Robust Multi-Scale Representation for NeRF from Unposed Images
  Nishant Jain, Suryansh Kumar, Luc Van Gool
  Internatinal Journal of Computer Vision (IJCV) 2024
  [ paper ] [ code ]
• How To Not Train Your Dragon
  Junting Chen, Guohao Li, Suryansh Kumar, Bernard Ghanem, F.Yu
  Robotics Science and Systems (RSS) 2023
  [ paper ] [ code ]
• Single Image Depth Prediction Made Better: A Multivariate Gaussian Take
  Ce Liu, Suryansh Kumar, Shuhang Gu, Radu Timofte, Luc Van Gool
  Conference on Computer Vision and Pattern Recognition (CVPR) 2023
  [ paper ] [ code ]
• Enhanced Stable View-Synthesis
  Nishant Jain*, Suryansh Kumar*, Luc Van Gool
  Conference on Computer Vision and Pattern Recognition (CVPR) 2023
  [ paper ] [ code ]
• VA-DepthNet: A Variational Approach to Single Image Depth Prediction
  Ce Liu, Suryansh Kumar, Shuhang Gu, Radu Timofte, Luc Van Gool
  International Conference on Learning Representations (ICLR) 2023 [Spot. Oral]
  [ paper ] [ code ]