Human Detection Phanindra Varma. Detection -- Overview  Human detection in static images is based on the HOG (Histogram of Oriented Gradients) encoding.

Slides:



Advertisements
Similar presentations
Object Recognition from Local Scale-Invariant Features David G. Lowe Presented by Ashley L. Kapron.
Advertisements

QR Code Recognition Based On Image Processing
Zhimin CaoThe Chinese University of Hong Kong Qi YinITCS, Tsinghua University Xiaoou TangShenzhen Institutes of Advanced Technology Chinese Academy of.
Histograms of Oriented Gradients for Human Detection
Human Identity Recognition in Aerial Images Omar Oreifej Ramin Mehran Mubarak Shah CVPR 2010, June Computer Vision Lab of UCF.
Chapter 8 Content-Based Image Retrieval. Query By Keyword: Some textual attributes (keywords) should be maintained for each image. The image can be indexed.
Computer Vision for Human-Computer InteractionResearch Group, Universität Karlsruhe (TH) cv:hci Dr. Edgar Seemann 1 Computer Vision: Histograms of Oriented.
Lecture 31: Modern object recognition
Many slides based on P. FelzenszwalbP. Felzenszwalb General object detection with deformable part-based models.
Histograms of Oriented Gradients for Human Detection Navneet Dalal and Bill Triggs CVPR 2005 Another Descriptor.
Activity Recognition Aneeq Zia. Agenda What is activity recognition Typical methods used for action recognition “Evaluation of local spatio-temporal features.
Presenter: Hoang, Van Dung
More sliding window detection: Discriminative part-based models Many slides based on P. FelzenszwalbP. Felzenszwalb.
DISCRIMINATIVE DECORELATION FOR CLUSTERING AND CLASSIFICATION ECCV 12 Bharath Hariharan, Jitandra Malik, and Deva Ramanan.
Edge Detection CSE P 576 Larry Zitnick
Patch Descriptors CSE P 576 Larry Zitnick
Generic Object Detection using Feature Maps Oscar Danielsson Stefan Carlsson
Ensemble Tracking Shai Avidan IEEE TRANSACTIONS ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE February 2007.
Multi-Class Object Recognition Using Shared SIFT Features
Object Detection using Histograms of Oriented Gradients
(1) Feature-point matching by D.J.Duff for CompVis Online: Feature Point Matching Detection, Extraction.
The SIFT (Scale Invariant Feature Transform) Detector and Descriptor
Fast Human Detection Using a Novel Boosted Cascading Structure With Meta Stages Yu-Ting Chen and Chu-Song Chen, Member, IEEE.
Lecture 6: Feature matching and alignment CS4670: Computer Vision Noah Snavely.
Scale-Invariant Feature Transform (SIFT) Jinxiang Chai.
Generic object detection with deformable part-based models
Exercise Session 10 – Image Categorization
Computer vision.
Watch, Listen and Learn Sonal Gupta, Joohyun Kim, Kristen Grauman and Raymond Mooney -Pratiksha Shah.
Object Detection with Discriminatively Trained Part Based Models
Lecture 31: Modern recognition CS4670 / 5670: Computer Vision Noah Snavely.
Pedestrian Detection and Localization
ECE738 Advanced Image Processing Face Detection IEEE Trans. PAMI, July 1997.
CSCE 643 Computer Vision: Extractions of Image Features Jinxiang Chai.
Handwritten Hindi Numerals Recognition Kritika Singh Akarshan Sarkar Mentor- Prof. Amitabha Mukerjee.
CS654: Digital Image Analysis Lecture 25: Hough Transform Slide credits: Guillermo Sapiro, Mubarak Shah, Derek Hoiem.
Lecture 7: Features Part 2 CS4670/5670: Computer Vision Noah Snavely.
COMP322/S2000/L171 Robot Vision System Major Phases in Robot Vision Systems: A. Data (image) acquisition –Illumination, i.e. lighting consideration –Lenses,
Histograms of Oriented Gradients for Human Detection(HOG)
Edge Detection and Geometric Primitive Extraction Jinxiang Chai.
Histograms of Oriented Gradients for Human Detection Navneet Dalal and Bill Triggs CVPR 2005 Another Descriptor.
Kylie Gorman WEEK 1-2 REVIEW. CONVERTING AN IMAGE FROM RGB TO HSV AND DISPLAY CHANNELS.
A Tutorial on using SIFT Presented by Jimmy Huff (Slightly modified by Josiah Yoder for Winter )
Week 10 Emily Hand UNR.
A New Method for Crater Detection Heather Dunlop November 2, 2006.
Scale Invariant Feature Transform (SIFT)
An ANN Approach to Identify if Driver is Wearing Safety Belts Hanwen Chen 12/9/2013.
Object Recognizing. Object Classes Individual Recognition.
A Statistical Approach to Texture Classification Nicholas Chan Heather Dunlop Project Dec. 14, 2005.
776 Computer Vision Jan-Michael Frahm Spring 2012.
More sliding window detection: Discriminative part-based models
Object Recognition Tutorial Beatrice van Eden - Part time PhD Student at the University of the Witwatersrand. - Fulltime employee of the Council for Scientific.
Week 4: 6/6 – 6/10 Jeffrey Loppert. This week.. Coded a Histogram of Oriented Gradients (HOG) Feature Extractor Extracted features from positive and negative.
Cascade for Fast Detection
Antoine Guitton, Geophysics Department, CSM
CS262: Computer Vision Lect 09: SIFT Descriptors
Scale Invariant Feature Transform (SIFT)
Data Driven Attributes for Action Detection
Lit part of blue dress and shadowed part of white dress are the same color
Recognition using Nearest Neighbor (or kNN)
Feature description and matching
Introduction of Pedestrian Detection
A Tutorial on HOG Human Detection
HOGgles Visualizing Object Detection Features
Categorization by Learning and Combing Object Parts
From a presentation by Jimmy Huff Modified by Josiah Yoder
Edge detection f(x,y) viewed as a smooth function
Edge Detection Today’s readings Cipolla and Gee Watt,
ECE734 Project-Scale Invariant Feature Transform Algorithm
Feature descriptors and matching
Presentation transcript:

Human Detection Phanindra Varma

Detection -- Overview  Human detection in static images is based on the HOG (Histogram of Oriented Gradients) encoding of images  Training set consists of positive windows (containing humans) and negative images  For each window in the training set the HOG feature vector is computed and linear SVM is used for learning the classifier  For any test image, the feature vector is computed on densely spaced windows at all scales and classified using the learned SVM

HOG encoding  Preprocessing:- Gamma normalize each channel using square root transformation in the given window For each channel compute gradients using [-1 0 1] and [-1 0 1] T and find the channel with the largest gradient magnitude for each pixel Compute gradient orientation (0 – 180) for each pixel in this dominant channel  Descriptor computation :- Divide the window (64x128) into dense grid of points with horizontal and vertical spacing equal to 8 pixels Divide the 16x16 region (block) centered on each point on the grid into cells of size 8x8 (i.e 4 cells for each grid point) For each pixel in the current block use Trilinear interpolation based on gradient strength to vote into a 2x2x9 histogram

HOG encoding (Contd..)  Different voting schemes were used for each of the colored regions  Block normalization for illumination invariance is done on each block independently using the norm of the 2x2x9 vector  The final feature vector is the collection of all the 2x2x9 feature vectors from all the grid points A Block of 16x16 pixels Cell centers Grid point

Training  The training set has been obtained from  The training set consists of positive 64x128 windows (2416) containing humans and negative images  Negative windows are sampled from the negative images at random locations (12000)  Initial Phase learning :- Learn the SVM classifier on the original training set  Generate Hard examples :- Run the learned SVM on the negative images at all scales and window locations and save all the false positives (approx.6000)

Training (Contd..)  Second Phase learning :- Using the newly generated negative examples learn the new linear SVM (total positive windows 2400, negative windows approx)  Following this procedure, 375 windows were misclassified out of the possible windows (using SVMLight)

Testing  Given an Image :- HOG feature vector is computed across all scales and window locations and the locations and scales of all positive windows are saved (window size 64x128)  This procedure gives multiple detections (at many scales and locations)  To fuse overlapping detections the Mean Shift mode detection algorithm is used  Represent each detection in a 3D space ([x y log(s)]) and iteratively compute the mean shift vector at each point  The resulting modes give the final detections and the bounding boxes are drawn using this final scale

Results - Detection An example image Detections when threshold is zero

Results – Detection (Contd..) Previous image Detections when threshold is equal to one

Results - Detection An example image Detections when threshold is zero

Results – Detection (Contd..) Result of Mean Shift mode detection

Comparision Detection Video