Should I Go to the Bathroom Now? Sketch of physics input from 6 - - PowerPoint PPT Presentation

should i go to the bathroom now
SMART_READER_LITE
LIVE PREVIEW

Should I Go to the Bathroom Now? Sketch of physics input from 6 - - PowerPoint PPT Presentation

Should I Go to the Bathroom Now? Sketch of physics input from 6 players on Kinect Me, playing with my balls Introduction Who am I? Chris Jurney! Introduction Who is Double Fine No, I will not buy you lunch Introduction


slide-1
SLIDE 1
slide-2
SLIDE 2

Should I Go to the Bathroom Now?

  • Sketch of physics input from 6 players on

Kinect

  • Me, playing with my balls
slide-3
SLIDE 3

Introduction

  • Who am I?

Chris Jurney!

slide-4
SLIDE 4
  • Who is Double Fine

Introduction

  • No, I will not buy you lunch
slide-5
SLIDE 5

Introduction

  • What is DFHAT?
slide-6
SLIDE 6

Introduction

  • Gameplay video…
slide-7
SLIDE 7

Introduction

  • Sizzle video
slide-8
SLIDE 8

The Problem

  • Motivation for Motion Blobs
slide-9
SLIDE 9

The Problem

  • Inputs (Kinect buffers)

Depth 6 Player Indices

slide-10
SLIDE 10

The Problem

  • Affecting the game
slide-11
SLIDE 11

The Solution

  • Problem with traditional vision solutions
slide-12
SLIDE 12

The Solution

  • Hacks to the rescue!
slide-13
SLIDE 13

Processing

1) Calculate motion 2) Filter motion c) Label image 4) Correlate labels over time to create blobs V) Create game inputs from blobs

slide-14
SLIDE 14

Calculate Motion

  • Compare to previous frame
  • Accept pixels new to the player &

player pixels with significant Z change

slide-15
SLIDE 15

Filter Motion

  • Blob shrink

– Clear noninterior pixels – Eliminates noise

  • Blob grow x 2

– Set pixels touching edge – Merges slightly disconnected areas

slide-16
SLIDE 16

Filter Motion

Motion Shrunk Grown x 2

slide-17
SLIDE 17

Label Image

  • For every untouched motion pixel

– Flood to all touching cells, setting the label – Calculate a 3D center of mass

slide-18
SLIDE 18

Motion & Label Video

slide-19
SLIDE 19
slide-20
SLIDE 20

Correlate Labels

  • Find new labels to match all the old labels
  • Biggest old blobs first
  • Axes of comparison:

– Player index – Center of mass vs. predicted position – Cell count

slide-21
SLIDE 21

Correlate Labels

  • If a match is found, update the blob

– Smooth out motion – Smooth out size

  • If a match is not found, ZOMBIE BLOB

– Extrapolate motion

  • Kill off old zombies
slide-22
SLIDE 22

Correlate Labels

  • Make new blobs for remaining big labels
slide-23
SLIDE 23

Gameplay Effects

  • Convert blobs to

collision spheres

  • Size is based on

blob area

  • Center is 3d center
  • f mass
  • Calculate gestures

from history

slide-24
SLIDE 24

Blob Debug Video

slide-25
SLIDE 25
slide-26
SLIDE 26

Pros

  • Efficient to calculate
  • Quick to implement
  • Action correlated to motion
  • Efficient collision representation with

velocities

slide-27
SLIDE 27

Cons

  • Only useful for crude gestures
  • Collision generated is innacurate
slide-28
SLIDE 28

Payoff Video

slide-29
SLIDE 29
slide-30
SLIDE 30