11-785 Introduction to Deep Learning
Spring 2020

Project Video Instruction

  1. Total video length: 3 - 5 minutes
  2. The video must be slides with voice over.
  3. Free softwares available for screen casting.
    1. Screen-o-matic: https://screencast-o-matic.com/screen-recorder
    2. Camtasia(free trial): https://www.techsmith.com/video-editor.html
    3. Others are also available
  4. Video must start with a slide of the title of the project, group name and names of the group members.
  5. Videos must contain enough information to be understood by someone not familiar with your work.

Contents:

1. Description of the problem(20 - 50 seconds)

  1. Description of the problem
  2. Significance of the problem
  3. Has the problem been solved before?If so, how good was the solution.

2. Task ( <= 30 seconds)

  1. Mention whether your work is re-implementation of a paper, application of an existing approach on a new dataset or a novel approach
  2. Dataset used, size of the dataset;show some examples from dataset in slides when possible

3. Approach( 1 minute - 2 minutes)

  1. Start with a brief description of the overall approach; Possibly include a diagram of your pipeline and/or architecture
  2. Describe the novel or most significant parts of your approach in greater detail; Use diagrams whenever possible: e.g. how your data is transformed, gradient flow in your model,...

4. Results and Discussion (<= 1 minute)

  1. Describe how performance was evaluated
  2. Report the best, interesting and unexpected results i.e no need to show the results for all your experiments
  3. Why did you get those results? Why did it work better/worse than other approaches?Possibly include diagrams comparing your work and others work.

5. Conclusions (<30 seconds)

General note:Resist the urge to describe everything in great detail to avoid going over the time limit