CS1674: Homework 10 - Written

Due: 4/17/2018, 11:59pm

This assignment is worth 10 points.
  1. [3 pts] What is the benefit of modeling the parts of an object and the spatial configuration of these parts, rather than modeling the object as a whole?
  2. [4 pts] List two differences between the R-CNN approach to object detection, and the Deformable Parts Model (DPM) approach.
  3. [3 pts] In what ways are action recognition at the frame level, and generating poetry, similar?