ecs 289h: visual recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf ·...
TRANSCRIPT
![Page 1: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/1.jpg)
ECS 289H: Visual RecognitionFall 2014
Yong Jae Lee
Department of Computer Science
![Page 2: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/2.jpg)
Plan for today
• Topic overview
• Introductions
• Course requirements and administrivia
• Syllabus tour (if time permits)
![Page 3: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/3.jpg)
Computer Vision
• Let computers see!
• Automatic understanding of visual data (i.e., images and video)
─ Computing properties of the 3D world from visual data (measurement)
─ Algorithms and representations to allow a machine to recognize objects, people, scenes, and activities (perception and interpretation)
Kristen Grauman
![Page 4: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/4.jpg)
What does recognition involve?
Fei-Fei Li
![Page 5: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/5.jpg)
Detection: are there people?
![Page 6: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/6.jpg)
Activity: What are they doing?
![Page 7: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/7.jpg)
Object categorization
mountain
building
tree
banner
vendor
people
street lamp
![Page 8: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/8.jpg)
Instance recognition
Potala
Palace
A particular
sign
![Page 9: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/9.jpg)
Scene categorization
• outdoor
• city
• …
![Page 10: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/10.jpg)
Attribute recognition
flat
gray
made of fabric
crowded
![Page 11: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/11.jpg)
Why recognition?
• Recognition a fundamental part of perception• e.g., robots, autonomous agents
• Organize and give access to visual content• Connect to information
• Detect trends and themes
• Why now?
Kristen Grauman
![Page 12: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/12.jpg)
Posing visual queries
Kooaba, Bay & Quack et al.
Yeh et al., MIT
Belhumeur et al.
Kristen Grauman
![Page 13: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/13.jpg)
Interactive systems
Shotton et al.
![Page 14: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/14.jpg)
Autonomous agents
Google self-driving car
Mars rover
![Page 15: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/15.jpg)
Exploring community photo collections
Snavely et al.
Simon & SeitzKristen Grauman
![Page 16: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/16.jpg)
Discovering visual patterns
Actions
Categories
Characteristic elements
Wang et al.
Doersch et al.
Lee & Grauman
![Page 17: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/17.jpg)
Recognizing flat, textured objects (e.g. books, CD covers, posters)
Reading license plates, zip codes, checks
Fingerprint recognition
Frontal face detection
What kinds of things work best today?
Kristen Grauman
![Page 18: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/18.jpg)
What are the challenges?
![Page 19: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/19.jpg)
What we see What a computer sees
![Page 20: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/20.jpg)
Challenges: robustness
Illumination Object pose
ViewpointIntra-class appearance
Occlusions
Clutter
Kristen Grauman
![Page 21: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/21.jpg)
Context cues
Kristen Grauman
Challenges: context and human experience
![Page 22: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/22.jpg)
Challenges: context and human experience
Context cues Function Dynamics
Kristen Grauman
![Page 23: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/23.jpg)
Challenges: context and human experience
Fei Fei Li, Rob Fergus, Antonio Torralba
![Page 24: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/24.jpg)
Challenges: context and human experience
Fei Fei Li, Rob Fergus, Antonio Torralba
![Page 25: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/25.jpg)
Challenges: context and human experience
Fei Fei Li, Rob Fergus, Antonio Torralba
![Page 26: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/26.jpg)
Challenges: scale, efficiency
• Half of the cerebral cortex in primates is devoted to processing visual information
![Page 27: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/27.jpg)
6 billion images 70 billion images 1 billion images served daily
10 billion images
100 hours uploaded per minute
Almost 90% of web traffic is visual!
:From
Challenges: scale, efficiency
![Page 28: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/28.jpg)
Biederman 1987
Challenges: scale, efficiency
Fei Fei Li, Rob Fergus, Antonio Torralba
![Page 29: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/29.jpg)
Challenges: learning with minimal supervision
MoreLess
Kristen Grauman
![Page 30: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/30.jpg)
Slide from Pietro Perona, 2004 Object Recognition workshop
![Page 31: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/31.jpg)
Slide from Pietro Perona, 2004 Object Recognition workshop
![Page 32: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/32.jpg)
Inputs in 1963…
L. G. Roberts, Machine Perception
of Three Dimensional Solids,
Ph.D. thesis, MIT Department of
Electrical Engineering, 1963.
Kristen Grauman
![Page 33: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/33.jpg)
Personal photo albums
Surveillance and security
Movies, news, sports
Medical and scientific images
… and inputs today
Svetlana Lazebnik
Understand and organize and index all this data!!
![Page 34: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/34.jpg)
![Page 35: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/35.jpg)
Introductions
![Page 36: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/36.jpg)
Course Information
• Website: https://sites.google.com/a/ucdavis.edu/ecs-289h-visual-recognition/
• Office hours: by appointment (email)
• Email: start subject with “[ECS289H]”
![Page 37: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/37.jpg)
This course
• Survey state-of-the-art research in computer vision
• High-level vision and learning problems
![Page 38: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/38.jpg)
Goals
• Understand, analyze, and discuss state-of-the-art techniques
• Identify interesting open questions and future directions
![Page 39: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/39.jpg)
Prerequisites
• Interest in computer vision!
• Courses in computer vision, machine learning, image processing is a plus (but not required)
![Page 40: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/40.jpg)
Requirements
• Class Participation [25%]
• Paper Reviews [25%]
• Paper Presentations (1-2 times) [25%]
• Final Project [25%]
![Page 41: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/41.jpg)
Class participation
• Actively participate in discussions
• Read all assigned papers before each class
![Page 42: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/42.jpg)
Paper reviews
• Detailed review of one paper before each class
• One page in length─ A summary of the paper (2-3 sentences)
─ Main contributions
─ Strengths and weaknesses
─ Experiments convincing?
─ Extensions?
─ Additional comments, including unclear points, open research questions, and applications
• Email by 11:59 pm the day before each class
• Email subject “[ECS 289H] paper review”
![Page 43: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/43.jpg)
Paper presentations
• 1 or 2 times during the quarter
• ~20 minutes long, well-organized, and polished:─ Clear statement of the problem
─ Motivate why the problem is interesting, important, and/or difficult
─ Describe key contributions and technical ideas
─ Experimental setup and results
─ Strengths and weaknesses
─ Interesting open research questions, possible extensions, and applications
• Slides should be clear and mostly visual (figures, animations, videos)
• Look at background papers as necessary
![Page 44: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/44.jpg)
Paper presentations
• Search for relevant material, e.g., from the authors' webpage, project pages, etc.
• Each slide that is not your own must be clearly cited
• Meet with me one week before your presentation with prepared slides
• Email slides on day of presentation
• Email subject "[ECS 289H] presentation slides"
• Skip paper review the day you are presenting
![Page 45: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/45.jpg)
Final project
• One of:─ Design and evaluation of a novel approach
─ An extension of an approach studied in class
─ In-depth analysis of an existing technique
─ In-depth comparison of two related existing techniques
• Work in pairs
• Talk to me if you need ideas
![Page 46: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/46.jpg)
Final project
• Final Project Proposal (5%) due 10/31─ 1 page
• Final Project Presentation (10%) due 12/9, 12/11─ 15 minutes
• Final Project Paper (10%) due TBD─ 6-8 pages
![Page 47: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/47.jpg)
![Page 48: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/48.jpg)
Local Features and Matching
• Local invariant features; detectors and descriptors
• Matching, indexing, bag-of-words
![Page 49: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/49.jpg)
Image classification
• Object and scene classification
• Global and object-based representations
![Page 50: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/50.jpg)
Object detection
• Discriminative methods
• Faces, people, objects
![Page 51: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/51.jpg)
Unsupervised/weakly-supervised discovery
• Clustering, context
• Mid-level visual elements
![Page 52: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/52.jpg)
Segmentation
• Grouping pixels
![Page 53: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/53.jpg)
Human-in-the-loop
• Interactive systems, crowdsourcing
• Active learning
![Page 54: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/54.jpg)
Activity recognition
• What are the humans in the image or video doing?
![Page 55: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/55.jpg)
Human pose
• Finding people and their poses
![Page 56: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/56.jpg)
Attributes
• Describing properties of objects, people, scenes
![Page 57: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/57.jpg)
Image search and mining
• Large-scale instance-based recognition
• Discovering connections
![Page 58: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/58.jpg)
Language and images
• Generating image descriptions
![Page 59: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/59.jpg)
Big data
• Data-driven techniques
• Dataset bias
![Page 60: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/60.jpg)
First-person vision
• Recognition from the first-person view
• Social scene understanding
![Page 61: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/61.jpg)
Deep convolutional neural networks
![Page 62: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/62.jpg)
![Page 63: ECS 289H: Visual Recognitionweb.cs.ucdavis.edu/~yjlee/teaching/ecs289h-fall2014/introduction.pdf · to recognize objects, people, scenes, and activities (perception and interpretation)](https://reader036.vdocuments.mx/reader036/viewer/2022071002/5fbf6468a0b8e2101e54b9bb/html5/thumbnails/63.jpg)
Coming up
• Carefully read class webpage
• Sign-up for papers you want to present
• Next class: overview of my research