Official Repository RVSS 2024

The material here provided was developed as part of the Robotic Vision Summer School.

---

Workshops

Please see this separate README.

---

Lectorials

You can run the notebooks via Colab: http://colab.research.google.com/github/rvss-australia/RVSS.

Links to slide decks and information about the lectorials will appear below.

---

Lectorial A: Robotic Vision

Presented by Peter Corke, Queensland University of Technology

The slides, details and resources for the A stream are here: A Slides

---

Lectorial B: Spatial Awareness

Presented by Teresa Vidal-Calleja, University of Technology Sydney

B1: Poses

Describing poses, transformations, and uncertainty

Slides:

[B1 Slides](coming soon)

Coding Sessions:

• Basic Geometry
• Uncertainty

B2: Uncertainty

Keeping track of stuff with imperfect measurement; Kalman filters, factor graphs, and batch optimisation

Slides:

[B2 Slides](coming soon)

Coding Sessions:

• Motion Model
• Kalman Filter 1D
• Multivariate Gaussian
• EKF
• SLAM

B3: Representations

A taxonomy of spatial representations

Slides:

[B3 Slides](coming soon)

---

Lectorial C: Visual Learning

Presented by Simon Lucey, University of Adelaide

C1: Convolutional Neural Networks

Introduction to convolutional neural networks

Slides:

[C1 Slides](coming soon)

Coding Session:

• Image classification with multi-layer perceptron

C2: Visual Transformers

Visual transformers and their connection to convolutional neural networks

Slides:

[C2](coming soon)

Coding Session:

• Image classification with Convolutional NN

C3: Point Clouds

Point cloud processing with transformers

Slides:

[C3 Slides](coming soon)

---

Lectorial D: Learning to Act

Presented by Dana Kulic, Monash University

D1: Learning from Demonstrations

Action selection as supervised learning; behaviour cloning

Slides:

• [D1 Slides](coming soon)

Coding Sessions:

• Introduction to Reinforcement Learning

D2: Learning from Experience

Introduction to reinforcement learning

Slides:

• [D2 Slides](coming soon)

Coding Sessions:

• Introduction to Model-Free Reinforcement Learning
• Deep RL - Replay Memory
• Deep RL - Target Network

D3: Inverse Reinforcement Learning

Introduction to inverse reinforcement learning

Slides:

[D3 Slides](coming soon)

Additional Resources

If you'd like to know more

• David Silver's RL Video Lectures at UCL
• Prof. Pascal Poupart's Video Lectures at University of Waterloo, Canada
• Sutton and Barton's Introduction to Reinforcement Learning book
• Sergey Levine's Video Lectures on deep reinforcement learning at UCBerkeley

---

Deep Dives

Stephen Gould, The Australian National University

• Slides
• Isaac Lecture Notes

Richard Hartley, The Australian National University

• Slides

Jen Jen Chung, The University of Queensland

• Slides