Sanja Fidler

Sanja Fidler

ning Research
  • Variational Amodal Object Completion
  • Learning Deformable Tetrahedral Meshes for 3D Reconstruction
  • Federated Simulation or Medical Imaging
  • Meta-Sim2: Unsupervised Learning of Scene Structure for Synthetic Data Generation
  • Lift, Splat, Shoot: Encoding Images from Arbitrary Camera Rigs by Implicitly Unprojecting to 3D
  • Learning to Simulate Dynamic Environments with GameGAN
  • Learning to Predict 3D Objects with an Interpolation-based Differentiable Renderer
  • Neural Turtle Graphics for Modeling City Road Layouts
  • Meta Sim: Learning to Generate Synthetic Datasets
  • Gated-SCNN Gated Shape CNNs for Semantic Segmentation
  • Devil is in the Edges: Learning Semantic Boundaries from Noisy Annotations