Assistive System Progress Report 1

Slides:



Advertisements
Similar presentations
Distinctive Image Features from Scale-Invariant Keypoints
Advertisements

Feature Detection. Description Localization More Points Robust to occlusion Works with less texture More Repeatable Robust detection Precise localization.
Distinctive Image Features from Scale-Invariant Keypoints David Lowe.
Object Recognition from Local Scale-Invariant Features David G. Lowe Presented by Ashley L. Kapron.
Author :Andrea Selinger Salgian Department of Computer Science
SIFT & MatLab Pier Luigi Mazzeo.
Matlab Tutorial. Session 2. SIFT
The SIFT (Scale Invariant Feature Transform) Detector and Descriptor
IBBT – Ugent – Telin – IPI Dimitri Van Cauwelaert A study of the 2D - SIFT algorithm Dimitri Van Cauwelaert.
Object Recognition with Invariant Features n Definition: Identify objects or scenes and determine their pose and model parameters n Applications l Industrial.
A Study of Approaches for Object Recognition
Object Recognition with Invariant Features n Definition: Identify objects or scenes and determine their pose and model parameters n Applications l Industrial.
SIFT on GPU ( the slides are not updated for newer versions of SiftGPU ) Changchang Wu 5/8/2007.
Distinctive Image Feature from Scale-Invariant KeyPoints
Distinctive image features from scale-invariant keypoints. David G. Lowe, Int. Journal of Computer Vision, 60, 2 (2004), pp Presented by: Shalomi.
Object Recognition Using Distinctive Image Feature From Scale-Invariant Key point D. Lowe, IJCV 2004 Presenting – Anat Kaspi.
Scale Invariant Feature Transform (SIFT)
Large Lump Detection by SVM Sharmin Nilufar Nilanjan Ray.
Distinctive Image Features from Scale-Invariant Keypoints David G. Lowe – IJCV 2004 Brien Flewelling CPSC 643 Presentation 1.
Scale-Invariant Feature Transform (SIFT) Jinxiang Chai.
Distinctive Image Features from Scale-Invariant Keypoints By David G. Lowe, University of British Columbia Presented by: Tim Havinga, Joël van Neerbos.
Internet-scale Imagery for Graphics and Vision James Hays cs195g Computational Photography Brown University, Spring 2010.
Bag of Visual Words for Image Representation & Visual Search Jianping Fan Dept of Computer Science UNC-Charlotte.
Object Tracking/Recognition using Invariant Local Features Applications l Mobile robots, driver assistance l Cell phone location or object recognition.
Reporter: Fei-Fei Chen. Wide-baseline matching Object recognition Texture recognition Scene classification Robot wandering Motion tracking.
CVPR 2003 Tutorial Recognition and Matching Based on Local Invariant Features David Lowe Computer Science Department University of British Columbia.
CSCE 643 Computer Vision: Extractions of Image Features Jinxiang Chai.
SIFT Algorithm Scale-invariant feature transform Extracts features that are robust to changes in image scale, noise, illumination, and local geometric.
Wenqi Zhu 3D Reconstruction From Multiple Views Based on Scale-Invariant Feature Transform.
Distinctive Image Features from Scale-Invariant Keypoints Ronnie Bajwa Sameer Pawar * * Adapted from slides found online by Michael Kowalski, Lehigh University.
Puzzle Solver Sravan Bhagavatula EE 638 Project Stanford ECE.
Kylie Gorman WEEK 1-2 REVIEW. CONVERTING AN IMAGE FROM RGB TO HSV AND DISPLAY CHANNELS.
Overview Introduction to local features Harris interest points + SSD, ZNCC, SIFT Scale & affine invariant interest point detectors Evaluation and comparison.
A Tutorial on using SIFT Presented by Jimmy Huff (Slightly modified by Josiah Yoder for Winter )
Scale Invariant Feature Transform (SIFT)
Augmented Reality and 3D modelling Done by Stafford Joemat Supervised by Mr James Connan.
Presented by David Lee 3/20/2006
CSCI 631 – Foundations of Computer Vision March 15, 2016 Ashwini Imran Image Stitching.
SIFT.
Speech Recognition through Neural Networks By Mohammad Usman Afzal Mohammad Waseem.
CSCI 631 – Foundations of Computer Vision March 15, 2016 Ashwini Imran Image Stitching Link: singhashwini.mesinghashwini.me.
Face recognition using Histograms of Oriented Gradients
10/1/ Cognitive Robotics1 Object Recognition Cognitive Robotics David S. Touretzky & Ethan Tira-Thompson Carnegie Mellon Spring 2008.
3D Puppetry: A Kinect-based Interface for 3D Animation
SIFT Scale-Invariant Feature Transform David Lowe
CS262: Computer Vision Lect 09: SIFT Descriptors
Presented by David Lee 3/20/2006
Lecture 07 13/12/2011 Shai Avidan הבהרה: החומר המחייב הוא החומר הנלמד בכיתה ולא זה המופיע / לא מופיע במצגת.
Distinctive Image Features from Scale-Invariant Keypoints
Scale Invariant Feature Transform (SIFT)
SIFT paper.
SIFT on GPU Changchang Wu 5/8/2007.
CS262 – Computer Vision Lect 08: SIFT Keypoint Detection
Car Recognition Through SIFT Keypoint Matching
EE 596 Machine Vision HW 6 Assigned: Nov 20, 2013
CAP 5415 Computer Vision Fall 2012 Dr. Mubarak Shah Lecture-5
Senior Design Capstone Project I
From a presentation by Jimmy Huff Modified by Josiah Yoder
The SIFT (Scale Invariant Feature Transform) Detector and Descriptor
SIFT.
Edge detection f(x,y) viewed as a smooth function
ECE734 Project-Scale Invariant Feature Transform Algorithm
Fourier Transform of Boundaries
SIFT SIFT is an carefully designed procedure with empirically determined parameters for the invariant and distinctive features.
Wavelet transform application – edge detection
Presented by Xu Miao April 20, 2005
THE ASSISTIVE SYSTEM SHIFALI KUMAR BISHWO GURUNG JAMES CHOU
Presentation transcript:

Assistive System Progress Report 1 Bishwo Gurung Shifali Kumar James Chou

Overview of the System Objective: To help individuals who are visually impaired and provide them independence. Object recognition system SIFT algorithm Speech recognition feature MATLAB (primary language and interactive environment)

SIFT - Scale-invariant feature transform Scale-space extrema detection Keypoint localization Orientation assignment Keypoint descriptor

The First Steps of Feature Detection Begin by constructing a scale space Replicate a natural concept of perception in digital imaging The scale space represents the image in multiple sizes to produce invariance.

Applying Gaussian Blur Intentionally eliminate unwanted detail from an image (such as textures or artifacts) Removes detail without introducing false ones by smoothing the image An important step in extracting keypoints

Combining with Octaves An octave contains a set of one specific image in the same scale but progressively blurs over each one. The next octave has the levels of blurring but is scaled half the size of the previous octave. This step is repeated over and over.

David Lowe’s ideal numbers (4 octaves, 5 blur levels)

Gaussian Blur Expression σ - scale parameter

Generate Difference of Gaussians new set of images for finding interest points.

Gaussian Filter blur_1 = imgaussfilt(I, sigma); blur_2 = imgaussfilt(blur_1, sigma); blur_3 = imgaussfilt(blur_2, sigma); blur_4 = imgaussfilt(blur_3, sigma);

Difference of Gaussians(Simplified) DoG_1 = I-blur_1; subplot(2,2,1); imshow(DoG_1) DoG_2 = blur_1-blur_2; subplot(2,2,2); imshow(DoG_1) DoG_3 = blur_2-blur_3; subplot(2,2,3); imshow(DoG_1) DoG_4 = blur_3-blur_4; subplot(2,2,4); imshow(DoG_1)

Sample Image

Next Step Key-point localization Orientation assignment Scale-space extrema detection Key-point localization Orientation assignment Keypoint descriptor Locate maxima/minima in DoG images

Speech Recognition There are 2 types of systems: isolated (having a brief pause between the words) continuous (talking fluently) Analyzing: speaker independent system (multiple people) speaker dependent system (single person)

Speech Recognition 3 stages main stages speech acquisition speech analysis user interface development We will be using the microphone built in the computer.

Speech Acquisition Matlab tool: Simulink library browser Training (part 1) This part is being done offline since we are training the system. First step, is to find a way to input in our data (word acoustic). Matlab tool: Simulink library browser Next, we need to know how the computer can differentiate between each of our inputs. A solution for this is creating our own acoustic model which will convert each sounds into numerical digits so it will be efficient for the system analyze.

Speech Acquisition Matlab tool: Data acquisition toolbox Testing (part 2) This part will be done will be live. We will need to pull the saved data and compare it with the new data while processing new speech. Matlab tool: Data acquisition toolbox Processing the new speech will be done frame by frame for isolated system.

Thank You!