Ivette Carreras Haroon Idrees

Slides:

Advertisements

Similar presentations

Figure 12.1 (a) Regions of brain activity in skilled readers during a visual feature detection (implicit reading) task; (b) a map of Brodmann areas (BAs)

Advertisements

Image Retrieval with Geometry-Preserving Visual Phrases

Location Recognition Given: A query image A database of images with known locations Two types of approaches: Direct matching: directly match image features.

Prep Year Curriculum What will my child learn?.

Three things everyone should know to improve object retrieval

Neurocomputing,Neurocomputing, Haojie Li Jinhui Tang Yi Wang Bin Liu School of Software, Dalian University of Technology School of Computer Science,

Computations have to be distributed !

Measurement. Measurements science is based on measurements all measurements have: – magnitude – uncertainty – units Numbers mathematics is based on numbers.

Stephan Gammeter, Lukas Bossard, Till Quack, Luc Van Gool.

CVPR 2008 James Philbin Ondˇrej Chum Michael Isard Josef Sivic

Effective Image Database Search via Dimensionality Reduction Anders Bjorholm Dahl and Henrik Aanæs IEEE Computer Society Conference on Computer Vision.

Histograms & Summary Data.  Summarizing large of amounts of data in two ways: Histograms: graphs give a pictorial representation of the data Numerical.

Information retrieval Finding relevant data using irrelevant keys Example: database of photographic images sorted by number, date. DBMS: Well structured.

Collaborative Recommendation via Adaptive Association Rule Mining KDD-2000 Workshop on Web Mining for E-Commerce (WebKDD-2000) Weiyang Lin Sergio A. Alvarez.

Salary Analytics™ Integrated salary reporting system raporty-placowe.pl Copyright by ProXmedia 2011.

From The World of Maps, by Judith A. Tyner. Copyright 2015 by The Guilford Press. Plate 6.1. Salton Sea photographed from Gemini 5, August 1965 (NASA photograph).

Scale Drawings & Scale Factor

 decimals/cc-7th-fracs-to-decimals/v/converting-fractions-to-decimals-example.

Beyond Sliding Windows: Object Localization by Efficient Subwindow Search The best paper prize at CVPR 2008.

Logan Lebanoff Mentor: Haroon Idrees. Two-layer method  Trying a method that will have two layers of neural networks.

12 Bar Blues. The notes in a key have Numbers abcdefgabcdefg abcdefgabcdefg

I. Problem  Improve large-scale retrieval / classification accuracy  Incorporate spatial relationship between the features in the image  Oxford 5K Dataset.

Hierarchical Matching with Side Information for Image Classification

Data Visualization Graphical representation of information Chars Plots Maps Time series.

Parts of a Map. Parts of a Map:  Most maps have the following six elements (parts), which help us read and understand what the map is trying to represent.

Parts of a Map.

Accessing Austin’s Census Data in Map Form Woo hoo!

Opinion Observer: Analyzing and Comparing Opinions on the Web

More on Document Similarity and Clustering How similar are these two documents (Again) ? Are these two documents about the same topic ?

Cost Curves Average Costs Marginal Costs Long run and Short Run.

Exercise 1 #include int main() { printf(“Hello C Programming!\n”); return 0; } 1.Run your Visual Studio 2008 or Create a new “project” and add.

A Scalable Association Rules Mining Algorithm Based on Sorting, Indexing and Trimming Chuang-Kai Chiou, Judy C. R Tseng Proceedings of the Sixth International.

1 The Strategies for Mining Fault-Tolerant Patterns Jia-Ling Koh Department of Information and Computer Education National Taiwan Normal University.

Template provided by MarketeersClub.com File Name MP_Competitive Summarys_0409 Competitive Summary – Competitors & Their Core Competency QualityServiceBenchmark.

Finding Text Trends l Word usage tracks interest changes l Segment documents by time period l Phrase frequency = number of documents l Phrase must have.

Enlargement Simple scale factors. Find the scale factor and the missing length ?

Biologically Inspired Vision-based Indoor Localization Zhihao Li, Ming Yang

Naifan Zhuang, Jun Ye, Kien A. Hua

Huffman Codes ASCII is a fixed length 7 bit code that uses the same number of bits to define each character regardless of how frequently it occurs. Huffman.

What I Did This Summer JEFFERSON RIDGEWAY 07 August 2017

Automatic Digitizing.

Hire Toyota Innova in Delhi for Outstation Tour

Supplementary Materials

Compact Query Term Selection Using Topically Related Text

Cheng-Ming Huang, Wen-Hung Liao Department of Computer Science

Counting in High-Density Crowd Videos

Counting in High-Density Crowd Videos

מדינת ישראל הוועדה לאנרגיה אטומית

How to Find Data Values (X) Given Specific Probabilities

I can draw dilations in the coordinate plane.

Enlargements and area Scale factor Area of object Area of image.

Place Physical Features Human Characteristics

REU Week 1 Ivette Carreras UCF.

Counting in High-Density Crowd Videos

Large Scale Image Deduplication

What will my child learn?

Manipulating and Sharing Data in a Database

Visual Communication Lab

ImageJ Introduction – Calculating distance and establishing scale using Google Maps and ImageJ Summary: You will be asked to calculate the length and width.

ENLARGEMENTS INTRODUCTION ENLARGEMENT Makes a shape bigger or smaller

Goal: The learner will use proportions to find measures of objects.

Computer-assisted alignment and tracking of acne lesions indicate that most inflammatory lesions arise from comedones and de novo Thy Thy Do, MBBS, Semyon.

DIRECTIONS: Click “New Slide” and pick the layout you want to use.

Counting in High-Density Crowd Videos

Ivette Carreras Haroon Idrees

CHALLENGE Match the Key phrases (A-J) with their Definitions (1-10)

Presentation transcript:

Ivette Carreras Haroon Idrees Visual Phrases Ivette Carreras Haroon Idrees

Selecting features to build phrases Experiment in landmark 1, 132 Images Features with high scale – top 50% Resulted in very short phrases: length 1-3 only All features regardless of scale Resulted in longer phrases: length 1-7

Steps to follow Go through every feature of every image and build the transactions Mine the resulting file Read and sort the found phrases by their frequency Select a percentage for the top frequency Currently using 20%

Steps to follow Go through all the features of every image and find a match for a given phrase Count the phrases in an image and build the Bag of Visual Phrases

Current Status Obtained transactions for all images Mined them with different minimum supports Min_support Length Sets found 25K 2 983 15K 3896 5K 3 4549 2.5K 5 6139 1.5K 8755 500 6 51985

Min_supt 500 Frequencies Phrases Length Frequency 42228 2 500-16354 1964 3 500-5166 179 4 500-3597 11 5 515-2728 1 6 558 Currently working on building the BoVP for these transactions

Visual Phrases using Data Mining for 132 Images. Phrase Length 2 38 29 37

Visual Phrases using Data Mining for 132 Images. Phrase Length 3 29 2

Next Steps Finish building the Bag of Visual Phrases for all images Find mAP for BoVP– mean average precision Compare results from BoW and BoVP Our 1K BoW – 20% mAP