3D Vision

Objective

The student will review and analyze the state-of-the-art techniques in three-dimensional computer vision, the objective is to understand the functioning and the scope of the subject in order to apply the best one for real-world problems related to computer engineering.

[Español] El estudiante revisará y analizará las técnicas más recientes en la visión computacional tridimensional con el objetivo de comprender el funcionamiento y alcance de las mismas en problemas reclacionados con la tecnología de cómputo.|

Requirements

• Obligatory
  • Linear algebra, Calculus,
  • Object oriented programming,
  • Statistics
• Desirable:
  • Introduction to computer vision
  • Pattern recognition.

Content

1. Introduction.
2. The vision in the real world.
3. Getting depth analytically
   1. Shape from X
   2. Stereo vision
   3. Multi-view geometry
4. Camera calibration
5. Monocular localization
   1. Feature points
   2. Kalman Filter
6. ORB slam
7. Deep Learning Review
   1. Intro
      https://youtu.be/Nzbthc0or_c
   2. Linear Regression
      https://youtu.be/1Z83VUtBda8
   3. Perceptron
   4. Multi-layer perceptron (MLP)
      https://youtu.be/Bb1Umcd7k-g
   5. Multinomial Logistic Classsifier
      https://youtu.be/qzHWRwGv9bw
   6. Convolutional neural networks (2 hrs)
8. Depth learning-based estimation (2 hrs)
9. 3D representations
   1. Introduction (2 hrs)
   2. Polyhedrons (2 hrs)
   3. Uniform grinds (2 hrs)
   4. Hierarchical grids (2 hrs)
   5. Probabilistic grids (2 hrs)
   6. Latent representations (2 hrs)
10. Measurements integration
    1. Registration (2 hrs)
    2. Bayes filter (2 hrs)
11. View Planning (2 hrs)
    1. Model based view planning (2 hrs)
    2. Next-best-view planning (2 hrs)
12. 3D object recognition
    1. CNN recognition (2 hrs)
    2. Next-best-view for recognition (2 hrs)

Bibliography

• Richard Hartley and Andrew Zisserman.Multiple View Geometry in Com-puter Vision. Cambridge University Press, 2 edition, 2004.
• Jeff Heaton. Ian goodfellow, yoshua bengio, and aaron courville: Deep learn-ing, 2018.
• Richard Szeliski.Computer vision: algorithms and applications. SpringerScience & Business Media, 2010.
• Sebastian Thrun. Probabilistic robotics.Communications of the ACM,45(3):52–57, 2002.

Papers

1. Niklaus, S., Mai, L., Yang, J., & Liu, F. (2019). 3D Ken Burns effect from a single image. ACM Transactions on Graphics (TOG), 38(6), 1-15.
2. Chen, R., Mahmood, F., Yuille, A., & Durr, N. J. (2018). Rethinking monocular depth estimation with adversarial training. arXiv preprint arXiv:1808.07528.