15
Realtime Object Recognition Using Decision Tree Learning Implemented with a Sony AIBO Robot CS 510 Presentation Chris Jorgensen

Realtime Object Recognition Using Decision Tree Learning

Embed Size (px)

DESCRIPTION

Realtime Object Recognition Using Decision Tree Learning. Implemented with a Sony AIBO Robot CS 510 Presentation Chris Jorgensen. Realtime Object Recognition Using Decision Tree Learning. Implemented with a Sony AIBO Robot CS 510 Presentation Chris Jorgensen. Presentation. - PowerPoint PPT Presentation

Citation preview

Page 1: Realtime Object Recognition Using Decision Tree Learning

Realtime Object Recognition Using Decision Tree Learning

Implemented with a Sony AIBO Robot

CS 510 Presentation

Chris Jorgensen

Page 2: Realtime Object Recognition Using Decision Tree Learning

Realtime Object Recognition Using Decision Tree Learning

Implemented with a Sony AIBO Robot

CS 510 Presentation

Chris Jorgensen

Page 3: Realtime Object Recognition Using Decision Tree Learning

Presentation

1. Abstract/Introduction

2. Problem setup

3. Use of decision tree learning

4. Results

5. Summary/Thoughts

Page 4: Realtime Object Recognition Using Decision Tree Learning

Abstract/Introduction

• Object recognition– Machine learning used to overcome issues:

• Domain-specific• Complexity inestimable• Quality of results

– Steps• Digital image scanned for features• Combine features into “meaningful” attributes• Attribute classification

Page 5: Realtime Object Recognition Using Decision Tree Learning

Introduction Continued…

Object Recognition Flow

Page 6: Realtime Object Recognition Using Decision Tree Learning

Preprocessing

• “Obvious” features– Colors– Limbs/Head

• Shapes derived from image– Used for feature

extraction

Page 7: Realtime Object Recognition Using Decision Tree Learning

Problem Setup

• Recognition– Iterate through surfaces

• Head, Side, Leg

– Generate segments for each surface– Store segments in memory

• 180 degree memory takes into account camera angle

Page 8: Realtime Object Recognition Using Decision Tree Learning

180 Degree Memory

Page 9: Realtime Object Recognition Using Decision Tree Learning

Problem Setup Continued

• Segmentation only done on “relevant” pixels– Determined by color

• Attribute generation*– Color, # segments, # corners, et al– Continuous values discretization via brute-

force generated optimal split

Page 10: Realtime Object Recognition Using Decision Tree Learning

Use of Decision Tree Learning

• Classification via Decision Tree Learning!– Algorithm creates a tree consisting of the

attributes; leafs are “symbols” • head, side, leg, body, et al

– Tree is built by calculating attribute with the highest entropy (depends on # occurrences of each value)

– Over-fitting solved by X2-pruning• Determine if attribute really detects a pattern

Page 11: Realtime Object Recognition Using Decision Tree Learning

Results

Page 12: Realtime Object Recognition Using Decision Tree Learning

Results Continued

Page 13: Realtime Object Recognition Using Decision Tree Learning

Results Continued

Page 14: Realtime Object Recognition Using Decision Tree Learning

Results Continued

• Decision Tree Learning– Classification (27 ms) “quite fast”– 84% precision on 1080 examples for 5

classes– Even a low number of examples (25) resulted

in over 50% precision– Room for improvement noted

Page 15: Realtime Object Recognition Using Decision Tree Learning

Summary/Thoughts

• Short/vague paper• Why do they need faster than 27 ms

recognition time? Aibos are slow!• Other work on Aibos done at PSU NWCIL

– Lendaris/Holmstrom– Aibo uses limb angles, model of motion, to

change gait based on floor surface– GA used to generate ideal gait for each

surface