Papers

13

Total Citations

318

H-Index

8

About

Sanja Fidler is a prominent AI researcher whose work spans 3D scene understanding, generative modeling, and human-robot interaction. Best known for her contributions to visual computing and simulation, Fidler has helped push the boundaries of how machines perceive, reconstruct, and synthesize complex environments. Her most impactful work includes physics-based human motion synthesis from video (77 citations), which eliminates the costly dependence on motion capture data, and NeuralField-LDM (43 citations), a hierarchical latent diffusion framework capable of generating richly detailed 3D environments — a breakthrough for virtual reality and robotics simulation. Her research into 3D Gaussian ray tracing (54 citations) advances real-time rendering of particle-based radiance fields, while GameGAN explores learning environment simulators directly from gameplay footage. Fidler has also made notable contributions to human-in-the-loop learning, developing systems that allow non-expert users to teach machines through natural language feedback — a vision deeply tied to accessible, household robotics. Her work on automated robot design via neural graph evolution further reflects her interdisciplinary reach across graphics, simulation, and autonomous systems. Across her career, Fidler has established herself as a leading voice in AI-driven 3D content creation and intelligent simulation.

Research Focus

Key Achievements

8
H-Index
13
Papers
318
Total Citations
24
Avg Citations/Paper
🏆 Most Cited Paper
Physics-based Human Motion Estimation and Synthesis from Videos
77 citations · 2021
📈 Most Prolific Year: 2021 (3 Papers)
🤝 Key Collaborators: 32
🏛 Institutions: Vector Institute, University of Toronto, University of Ljubljana, Nvidia (United Kingdom)

Top Papers

  1. 1
  2. 2
  3. 3
  4. 4
  5. 5
  6. 6
  7. 7
  8. 8
  9. 9
  10. 10

Key Collaborators

Contact & Links

Available for collaboration
Content generated · 14 days ago