Learning-based depth estimation from stereo and monocular images: successes, limitations and future challenges

3DV 2018 Tutorial: Learning-based depth estimation from stereo and monocular images: successes, limitations and future challenges

Location: University of Verona, Strada le Grazie 15, Verona (see below)

Date: Saturday, September 8, 2018

Time: 9.00-17.30 (Full day)

Organizers

Postdoctoral Fellow, Department of Computer Science and Engineering, University of Bologna

PhD student, Department of Computer Science and Engineering, University of Bologna

PhD student, Department of Computer Science , Stevens Institute of Technology

Associate Professor, Department of Computer Science , Stevens Institute of Technology

Associate Professor, Department of Computer Science and Engineering, University of Bologna

Schedule

Morning session:

  • Introduction and welcome (Organizers)
  • Confidence measures and early machine-learning for stereo (S. Mattoccia) [PDF]
  • Learning-based matching functions (K. Batsos) [PDF]
  • Learning for optimization and post-processing (K. Batsos and P. Mordohai) [PDF]
  • End-to-end stereo (M. Poggi) [PDF]

Afternoon session:

  • Learning for multi-view stereo (P. Mordohai) [PDF]
  • End-to-end unsupervised monocular depth estimation (F. Tosi) [PDF]
  • Adaptation techniques (M. Poggi) [PDF]
  • Discussion