Modeling Complex Temporal Composition of Actionlets for Activity Prediction

Size: px
Start display at page:

Download "Modeling Complex Temporal Composition of Actionlets for Activity Prediction"

Transcription

1 Modeling Complex Temporal Composition of Actionlets for Activity Prediction ECCV 2012 Activity Recognition Reading Group

2 Framework of activity prediction

3 What is an Actionlet To segment a long sequence of human activity video into multiple atomic actions that have specific semantic meanings. These atomic actions have causal relationships that can be used for prediction. We call these atomic actions actionlets.

4 Datasets Requirement: the activity should have multiple steps where each step constitutes a meaningful action unit (actionlet). Two datasets: Tennis games from YouTube. Maryland Human-Object Interaction (MHOI) dataset.

5 Dataset: Tennis Game 160 video clips.(80 winning, 80 losing). Each video clip is one point with an exchange of several strokes. Prediction problem: who will win?

6 Dataset: Maryland Human-Object Interaction Five activity categories: Answering phone call Making phone call Drinking water Lighting a flashlight Pouring water

7 Dataset: Maryland Human-Object Interaction Five activity categories. These activities have about 3 to 5 actionlets each. Constituent actionlets share similar human movements: 1) reaching for an object 2) grasping the object 3) manipulating the object, and 4) put back the object. 44 video clips. Prediction problem: what action is it?

8 Algorithm outline Given a set of training videos: Step 1: Actionlets Detection Segment each video into actionlets. Step 2: Activity Encoding Cluster actionlets into meaningful groups. Step 3: Activity Prediction Model Generate a next symbol probability function for each activity class. Given an ongoing test sequence: Compute the probability of the sequence under each prediction models, select the largest one as the predicted class.

9 Step 1: Actionlet Detection Input: a set of videos. Goal: find the frame indices that can segment the video into actionlets. Method: velocity based. For each atomic action, the start frame and the end frame always have the lowest movement velocity. And the velocity reaches the peak at the intermediate stage of each action.

10 Step 1: Actionlet Detection Input: a set of videos. Goal: find the frame indices that can segment the video into actionlets. Method: velocity based. For each atomic action, the start frame and the end frame always have the lowest movement velocity. And the velocity reaches the peak at the intermediate stage of each action.

11 Step 1: Actionlet Detection Input: a set of videos. Goal: find the frame indices that can segment the video into actionlets. Method: velocity based. S1. Key points: Harris Corner S2. Trajectories: Lucas-Kanade optical flow S3: Generate overall motion velocity V t for each frame F t :

12 Step 1: Actionlet Detection Result Evaluation Evaluation: Manually label the segmentation point in the video. A target window with size of 15 frames around the human labeled segmentation point would be the ground truth. MHOI: Manually labeled 113 actionlets for all 44 videos. Accuracy is Tennis: Manually labeled 253 actionlets for 40(out of 160) videos. Accuracy is 0.82.

13 Step 2: Activity Encoding Input: a set of segmented videos. Goal: Cluster actionlet into meaningful groups, then represent each video by a sequence of actionlets. Method: S1. Compute video descriptors for each actionlet; S2. Quantize the descriptors by k-means; S3. Learn the actionlet categories.

14 Step 2: Activity Encoding S1: Video descriptors Input: a set of segmented videos. MHOI: large scale person + static background Interest points: 3-D Harris corner. Descriptor: HOG and HOF. Tennis: small scale person Dense trajectories [18]. Feature points in dense grid -> Trajectories by optical flow + median filtering -> HOG, HOF and MBH features of each trajectories. [18]H. Wang, A. Klaser, C. Schmid, C.L. Liu, Action recognitino by dense trajectories.

15 Step 2: Activity Encoding S2: Descriptor quantization Input: actionlets represented as descriptors. 1. Cluster all descriptors by k-means; 2. Generate the descriptor codebook; 3. Compute memberships with respect to the codebook. Codebook size: MHOI: 50. Tennis: 1000.

16 Step 2: Activity Encoding S3: Actionlet categories Input: Actionlets represented as a collection of words from the codebook (word histogram). Learn the actionlets categories from the histograms in an unsupervised way [17]. [17] J. Niebles, H.Wang, L. Fei-Fei, Unsupervised Learning of Hunan Action Categories Using Spatial-Temporal Words.

17 Step 2: Activity Encoding Input: a set of segmented videos. Goal: Cluster actionlet into meaningful groups, then represent each video by a sequence of actionlets. Method: S1. Compute video descriptors for each actionlet; S2. Quantize the descriptors by k-means; S3. Learn the actionlet categories. Evaluation

18 Step 2: Activity Encoding Result Evaluation Evaluate the actionlets clustering result. Measurement: Rand Index. A set of elements S, two partitions X and Y. Define: a: the number of pairs of elements in S that are in the same set in X and in the same set in Y; b: different sets in X and in different sets in Y; c: same set in X and in different sets in Y; d: different sets in X and in the same set in Y. R = (a + b) / (a + b + c + d). The value is between 0 and 1, with 1 denoting two clusters are exactly the same.

19 Step 2: Activity Encoding Result Evaluation Evaluate the actionlets clustering result. Measurement: Rand Index. MHOI: Manually group actionlets into 5 categories where each category represents a meaningful action unit, e.g. reach the object, grab the object, release the object. Rand index is 0.7. Tennis: Group 253 actionlets from 40 videos into 10 categories. Rand index is 0.73.

20 Step 3: Activity Encoding Input: videos represented by actionlet sequences. Σ : actionlet alphabet. training 1 2 m {,,..., }:, D = r r r training sample set where r = r r... r, r Σ. i i i i i 1 2 l j i Goal: to learn a model P (for each activity) that provides a probability assignment p(t) for the ongoing actionlet sequence t = t1, t2,..., t t. Tool: Probabilistic Suffix Tree (PST)

21 Step 3: Activity Encoding Probabilistic Suffix Tree (PST) Given training data D training of an activity, Capture the temporal dependency between actionlets (next symbol probability function). The dependency is modeled as a Variable order Markov Model (VMM). Represented by Probabilistic Suffix Tree.

22 Step 3: Activity Encoding Probabilistic Suffix Tree (PST) next symbol probability function : γ ( σ), where σ Σ s Σ context *, ( ). s

23 Step 3: Activity Encoding Probabilistic Suffix Tree (PST)

24 Step 3: Activity Encoding Input: videos represented by actionlet sequences. Σ : actionlet alphabet. D = r r r training sample set training 1 2 m {,,..., }:, where r = r r... r, r Σ. i i i i i 1 2 l j i Goal: to learn a model P for each activity that provides a probability assignment p(t) for the ongoing actionlet sequence t = t1, t2,..., t t. Tool: Probabilistic Suffix Tree (PST) t T p ( t) = log γ j 1 ( tj ) j= 1 s

25 Step 3: Activity Encoding Predictive Accumulative Function(PAF) Predictability of different activities. Observations: Tennis game: late predictable. Only the last several strokes impact the winning or losing results. Drinking water : early predictable. As long as the first actionlet grab a cup is observed, we can guess the intention. How to use: weighted log likelihood. p tt... t t T 1 2 j ( t) = f p log γ j 1 ( t j ) s j= 1 t

26 Step 3: Activity Encoding Predictive Accumulative Function(PAF) Predictability of different activities. Observations: Tennis game: late predictable. Only the last several strokes impact the winning or losing results. Drinking water : early predictable. As long as the first actionlet grab a cup is observed, we can guess the intention. How to use: weighted log likelihood. p tt... t t T 1 2 j ( t) = f p log γ j 1 ( t j ) s j= 1 t

27 Step 3: Activity Encoding Predictive Accumulative Function(PAF) Notations: for k [0,1]. 1 2 m i i i i D = { r, r,..., r }, where r = r r... r. With the learnt PST we can compute: Given the first k percentage of the sequence observed, the information we gain is: T T H( D) H( D D ) H( D) = p ()log r p () r Fit a function: r rpre( k ) T γ j 1 ( ) j 1 s j pre k γ s pre( k ) j= 1 j= D = r r where r = r r r 1 m i i i i k { pre( k ),..., pre( k )}, pre( k ) k l. T p ( r) = ( r ), p ( r ) = ( r ) y k = H( D) k, y = f ( ) p k l i i r D T T H( D Dk ) = p ( r, rpre( k )) log p ( r rpre( k )) rpre( k ) Dk r D

28 Step 3: Activity Encoding Input: videos represented by actionlet sequences. Σ : actionlet alphabet. D = r r r training sample set training 1 2 m {,,..., }:, where r = r r... r, r Σ. i i i i i 1 2 l j i Goal: to learn a model P for each activity that provides a probability assignment p(t) for the ongoing actionlet sequence t = t1, t2,..., t t. Weighted log likelihood p tt... t t T 1 2 j ( t) = f p log γ j 1 ( t j ) s j= 1 t

29 Algorithm outline Given a set of training videos: Step 1: Actionlets Detection Segment each video into actionlets. Step 2: Activity Encoding Cluster actionlets into meaningful groups. Step 3: Activity Prediction Model Generate a next symbol probability function for each activity class. T Given an ongoing test sequence: () Compute the probability of the sequence under each prediction models, select the largest one as the predicted class. p t for each activity

30 Experiment Results Middle-Level Complex Activity Prediction: On MHOI High-Level Complex Activity Prediction: On Tennis

31 Experiment Results Middle-Level Complexity (MHOI) Dataset: 5 activities, each has 9 or 10 samples. Supervised classification problem: For each activity: Positive set: all the samples of this activity. Negative set: equal number of samples randomly selected from other activities. Evaluate accuracy by leave-one-out. Repeat above for 10 times, and average the performance. Other methods for comparison: Dynamic Bag-of-Words Integral Bag-of-Words A basic SVM-based approach

32 Experiment Results Middle-Level Complexity (MHOI)

33 Experiment Results Middle-Level Complexity (MHOI)

34 Experiment Results Middle-Level Complexity (MHOI)

35 Experiment Results High-Level Complexity (Tennis) Dataset: 2 categories (winning/losing), each has 80 samples. Length of actionlet sequences vary from 1 to more than 20. Supervised classification problem: Evaluate accuracy by leave-one-out. Other methods for comparison: Dynamic Bag-of-Words Integral Bag-of-Words A basic SVM-based approach

36 Experiment Results High-Level Complexity (Tennis)

37 Experiment Results High-Level Complexity (Tennis)

38

Global Scene Representations. Tilke Judd

Global Scene Representations. Tilke Judd Global Scene Representations Tilke Judd Papers Oliva and Torralba [2001] Fei Fei and Perona [2005] Labzebnik, Schmid and Ponce [2006] Commonalities Goal: Recognize natural scene categories Extract features

More information

Histogram of multi-directional Gabor filter bank for motion trajectory feature extraction

Histogram of multi-directional Gabor filter bank for motion trajectory feature extraction Histogram of multi-directional Gabor filter bank for motion trajectory feature extraction NGOC NAM BUI, TAN DAT TRINH, MIN KYUNG PARK, JIN YOUNG KIM Dept. of Electronics and Computer Engineering Chonnam

More information

Distinguish between different types of scenes. Matching human perception Understanding the environment

Distinguish between different types of scenes. Matching human perception Understanding the environment Scene Recognition Adriana Kovashka UTCS, PhD student Problem Statement Distinguish between different types of scenes Applications Matching human perception Understanding the environment Indexing of images

More information

Representing Visual Appearance by Video Brownian Covariance Descriptor for Human Action Recognition

Representing Visual Appearance by Video Brownian Covariance Descriptor for Human Action Recognition Representing Visual Appearance by Video Brownian Covariance Descriptor for Human Action Recognition Piotr Bilinski, Michal Koperski, Slawomir Bak, Francois Bremond INRIA 2004 Route des Lucioles, BP 93,

More information

Anticipating Visual Representations from Unlabeled Data. Carl Vondrick, Hamed Pirsiavash, Antonio Torralba

Anticipating Visual Representations from Unlabeled Data. Carl Vondrick, Hamed Pirsiavash, Antonio Torralba Anticipating Visual Representations from Unlabeled Data Carl Vondrick, Hamed Pirsiavash, Antonio Torralba Overview Problem Key Insight Methods Experiments Problem: Predict future actions and objects Image

More information

Learning Spa+otemporal Graphs of Human Ac+vi+es

Learning Spa+otemporal Graphs of Human Ac+vi+es Learning Spa+otemporal Graphs of Human Ac+vi+es William Brendel Sinisa Todorovic Our Goal Long Jump Triple Jump Recognize all occurrences of activities Identify the start and end frames Parse the video

More information

Statistical Filters for Crowd Image Analysis

Statistical Filters for Crowd Image Analysis Statistical Filters for Crowd Image Analysis Ákos Utasi, Ákos Kiss and Tamás Szirányi Distributed Events Analysis Research Group, Computer and Automation Research Institute H-1111 Budapest, Kende utca

More information

Human activity recognition in the semantic simplex of elementary actions

Human activity recognition in the semantic simplex of elementary actions BEAUDRY et al.: ACTIVITY RECOGNITION IN THE SEMANTIC SIMPLEX 1 Human activity recognition in the semantic simplex of elementary actions Beaudry Cyrille cyrille.beaudry@univ-lr.fr Péteri Renaud renaud.peteri@univ-lr.fr

More information

Deep Reinforcement Learning for Unsupervised Video Summarization with Diversity- Representativeness Reward

Deep Reinforcement Learning for Unsupervised Video Summarization with Diversity- Representativeness Reward Deep Reinforcement Learning for Unsupervised Video Summarization with Diversity- Representativeness Reward Kaiyang Zhou, Yu Qiao, Tao Xiang AAAI 2018 What is video summarization? Goal: to automatically

More information

Kai Yu NEC Laboratories America, Cupertino, California, USA

Kai Yu NEC Laboratories America, Cupertino, California, USA Kai Yu NEC Laboratories America, Cupertino, California, USA Joint work with Jinjun Wang, Fengjun Lv, Wei Xu, Yihong Gong Xi Zhou, Jianchao Yang, Thomas Huang, Tong Zhang Chen Wu NEC Laboratories America

More information

Global Behaviour Inference using Probabilistic Latent Semantic Analysis

Global Behaviour Inference using Probabilistic Latent Semantic Analysis Global Behaviour Inference using Probabilistic Latent Semantic Analysis Jian Li, Shaogang Gong, Tao Xiang Department of Computer Science Queen Mary College, University of London, London, E1 4NS, UK {jianli,

More information

Towards Good Practices for Action Video Encoding

Towards Good Practices for Action Video Encoding Towards Good Practices for Action Video Encoding Jianxin Wu National Key Laboratory for Novel Software Technology Nanjing University, China wujx21@nju.edu.cn Yu Zhang Nanyang Technological University Singapore

More information

Anomaly Detection for the CERN Large Hadron Collider injection magnets

Anomaly Detection for the CERN Large Hadron Collider injection magnets Anomaly Detection for the CERN Large Hadron Collider injection magnets Armin Halilovic KU Leuven - Department of Computer Science In cooperation with CERN 2018-07-27 0 Outline 1 Context 2 Data 3 Preprocessing

More information

Clustering with k-means and Gaussian mixture distributions

Clustering with k-means and Gaussian mixture distributions Clustering with k-means and Gaussian mixture distributions Machine Learning and Category Representation 2012-2013 Jakob Verbeek, ovember 23, 2012 Course website: http://lear.inrialpes.fr/~verbeek/mlcr.12.13

More information

Style-aware Mid-level Representation for Discovering Visual Connections in Space and Time

Style-aware Mid-level Representation for Discovering Visual Connections in Space and Time Style-aware Mid-level Representation for Discovering Visual Connections in Space and Time Experiment presentation for CS3710:Visual Recognition Presenter: Zitao Liu University of Pittsburgh ztliu@cs.pitt.edu

More information

Anomaly Localization in Topic-based Analysis of Surveillance Videos

Anomaly Localization in Topic-based Analysis of Surveillance Videos Anomaly Localization in Topic-based Analysis of Surveillance Videos Deepak Pathak IIT Kanpur Dept. of Computer Science deepakp@iitk.ac.in Abstract Abhijit Sharang IIT Kanpur Dept. of Computer Science abhisg@iitk.ac.in

More information

from Object Image Based on

from Object Image Based on Inference of Grasping Pattern from Object Image Based on Interaction Descriptor Tadashi Matsuo, Takuya Kawakami, Yoko Ogawa, Nobutaka Shimada Ritsumeikan University Introduction An object as a tool has

More information

Urban land use information retrieval based on scene classification of Google Street View images

Urban land use information retrieval based on scene classification of Google Street View images Urban land use information retrieval based on scene classification of Google Street View images Xiaojiang Li 1, Chuanrong Zhang 1 1 Department of Geography, University of Connecticut, Storrs Email: {xiaojiang.li;chuanrong.zhang}@uconn.edu

More information

Wavelet-based Salient Points with Scale Information for Classification

Wavelet-based Salient Points with Scale Information for Classification Wavelet-based Salient Points with Scale Information for Classification Alexandra Teynor and Hans Burkhardt Department of Computer Science, Albert-Ludwigs-Universität Freiburg, Germany {teynor, Hans.Burkhardt}@informatik.uni-freiburg.de

More information

Pattern Recognition and Machine Learning. Learning and Evaluation of Pattern Recognition Processes

Pattern Recognition and Machine Learning. Learning and Evaluation of Pattern Recognition Processes Pattern Recognition and Machine Learning James L. Crowley ENSIMAG 3 - MMIS Fall Semester 2016 Lesson 1 5 October 2016 Learning and Evaluation of Pattern Recognition Processes Outline Notation...2 1. The

More information

SYMBOL RECOGNITION IN HANDWRITTEN MATHEMATI- CAL FORMULAS

SYMBOL RECOGNITION IN HANDWRITTEN MATHEMATI- CAL FORMULAS SYMBOL RECOGNITION IN HANDWRITTEN MATHEMATI- CAL FORMULAS Hans-Jürgen Winkler ABSTRACT In this paper an efficient on-line recognition system for handwritten mathematical formulas is proposed. After formula

More information

Brief Introduction of Machine Learning Techniques for Content Analysis

Brief Introduction of Machine Learning Techniques for Content Analysis 1 Brief Introduction of Machine Learning Techniques for Content Analysis Wei-Ta Chu 2008/11/20 Outline 2 Overview Gaussian Mixture Model (GMM) Hidden Markov Model (HMM) Support Vector Machine (SVM) Overview

More information

Hidden CRFs for Human Activity Classification from RGBD Data

Hidden CRFs for Human Activity Classification from RGBD Data H Hidden s for Human from RGBD Data Avi Singh & Ankit Goyal IIT-Kanpur CS679: Machine Learning for Computer Vision April 13, 215 Overview H 1 2 3 H 4 5 6 7 Statement H Input An RGBD video with a human

More information

Human Action Recognition under Log-Euclidean Riemannian Metric

Human Action Recognition under Log-Euclidean Riemannian Metric Human Action Recognition under Log-Euclidean Riemannian Metric Chunfeng Yuan, Weiming Hu, Xi Li, Stephen Maybank 2, Guan Luo, National Laboratory of Pattern Recognition, Institute of Automation, CAS, Beijing,

More information

A Human Behavior Recognition Method Based on Latent Semantic Analysis

A Human Behavior Recognition Method Based on Latent Semantic Analysis Journal of Information Hiding and Multimedia Signal Processing c 2016 ISSN 2073-4212 Ubiquitous International Volume 7, Number 3, May 2016 A Human Behavior Recognition Method Based on Latent Semantic Analysis

More information

Shared Segmentation of Natural Scenes. Dependent Pitman-Yor Processes

Shared Segmentation of Natural Scenes. Dependent Pitman-Yor Processes Shared Segmentation of Natural Scenes using Dependent Pitman-Yor Processes Erik Sudderth & Michael Jordan University of California, Berkeley Parsing Visual Scenes sky skyscraper sky dome buildings trees

More information

Convolutional Neural Networks

Convolutional Neural Networks Convolutional Neural Networks Books» http://www.deeplearningbook.org/ Books http://neuralnetworksanddeeplearning.com/.org/ reviews» http://www.deeplearningbook.org/contents/linear_algebra.html» http://www.deeplearningbook.org/contents/prob.html»

More information

Compressed Fisher vectors for LSVR

Compressed Fisher vectors for LSVR XRCE@ILSVRC2011 Compressed Fisher vectors for LSVR Florent Perronnin and Jorge Sánchez* Xerox Research Centre Europe (XRCE) *Now with CIII, Cordoba University, Argentina Our system in a nutshell High-dimensional

More information

Loss Functions and Optimization. Lecture 3-1

Loss Functions and Optimization. Lecture 3-1 Lecture 3: Loss Functions and Optimization Lecture 3-1 Administrative Assignment 1 is released: http://cs231n.github.io/assignments2017/assignment1/ Due Thursday April 20, 11:59pm on Canvas (Extending

More information

Mining Motion Atoms and Phrases for Complex Action Recognition

Mining Motion Atoms and Phrases for Complex Action Recognition Mining Motion Atoms and Phrases for Complex Action Recognition LiMin Wang 1,2, Yu Qiao 2, and Xiaoou Tang 1,2 1 Department of Information Engineering, The Chinese University of Hong Kong 2 Shenzhen key

More information

What is semi-supervised learning?

What is semi-supervised learning? What is semi-supervised learning? In many practical learning domains, there is a large supply of unlabeled data but limited labeled data, which can be expensive to generate text processing, video-indexing,

More information

Predictive analysis on Multivariate, Time Series datasets using Shapelets

Predictive analysis on Multivariate, Time Series datasets using Shapelets 1 Predictive analysis on Multivariate, Time Series datasets using Shapelets Hemal Thakkar Department of Computer Science, Stanford University hemal@stanford.edu hemal.tt@gmail.com Abstract Multivariate,

More information

Beyond Spatial Pyramids

Beyond Spatial Pyramids Beyond Spatial Pyramids Receptive Field Learning for Pooled Image Features Yangqing Jia 1 Chang Huang 2 Trevor Darrell 1 1 UC Berkeley EECS 2 NEC Labs America Goal coding pooling Bear Analysis of the pooling

More information

Multimodal context analysis and prediction

Multimodal context analysis and prediction Multimodal context analysis and prediction Valeria Tomaselli (valeria.tomaselli@st.com) Sebastiano Battiato Giovanni Maria Farinella Tiziana Rotondo (PhD student) Outline 2 Context analysis vs prediction

More information

Speech-driven Facial Animation

Speech-driven Facial Animation peech-driven Facial Animation how to learn a stream-to-stream mapping? Hiroshi himodaira (ICC, CTR) Junichi Yamagishi, Gregor Hofer, ichael Berger peech-driven facial animation? It's a computer animated

More information

When Dictionary Learning Meets Classification

When Dictionary Learning Meets Classification When Dictionary Learning Meets Classification Bufford, Teresa 1 Chen, Yuxin 2 Horning, Mitchell 3 Shee, Liberty 1 Mentor: Professor Yohann Tendero 1 UCLA 2 Dalhousie University 3 Harvey Mudd College August

More information

Unsupervised Activity Perception in Crowded and Complicated Scenes Using Hierarchical Bayesian Models

Unsupervised Activity Perception in Crowded and Complicated Scenes Using Hierarchical Bayesian Models SUBMISSION TO IEEE TRANS. ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 1 Unsupervised Activity Perception in Crowded and Complicated Scenes Using Hierarchical Bayesian Models Xiaogang Wang, Xiaoxu Ma,

More information

University of Genova - DITEN. Smart Patrolling. video and SIgnal Processing for Telecommunications ISIP40

University of Genova - DITEN. Smart Patrolling. video and SIgnal Processing for Telecommunications ISIP40 University of Genova - DITEN Smart Patrolling 1 Smart Patrolling Detection of the intruder Tracking of the intruder A cognitive node will active an operator, describing on his mobile terminal the characteristic

More information

Mathematical Formulation of Our Example

Mathematical Formulation of Our Example Mathematical Formulation of Our Example We define two binary random variables: open and, where is light on or light off. Our question is: What is? Computer Vision 1 Combining Evidence Suppose our robot

More information

Data Preprocessing. Cluster Similarity

Data Preprocessing. Cluster Similarity 1 Cluster Similarity Similarity is most often measured with the help of a distance function. The smaller the distance, the more similar the data objects (points). A function d: M M R is a distance on M

More information

Video and Motion Analysis Computer Vision Carnegie Mellon University (Kris Kitani)

Video and Motion Analysis Computer Vision Carnegie Mellon University (Kris Kitani) Video and Motion Analysis 16-385 Computer Vision Carnegie Mellon University (Kris Kitani) Optical flow used for feature tracking on a drone Interpolated optical flow used for super slow-mo optical flow

More information

Data-Intensive Computing with MapReduce

Data-Intensive Computing with MapReduce Data-Intensive Computing with MapReduce Session 8: Sequence Labeling Jimmy Lin University of Maryland Thursday, March 14, 2013 This work is licensed under a Creative Commons Attribution-Noncommercial-Share

More information

Lecture 13 Visual recognition

Lecture 13 Visual recognition Lecture 13 Visual recognition Announcements Silvio Savarese Lecture 13-20-Feb-14 Lecture 13 Visual recognition Object classification bag of words models Discriminative methods Generative methods Object

More information

Probabilistic Latent Semantic Analysis

Probabilistic Latent Semantic Analysis Probabilistic Latent Semantic Analysis Dan Oneaţă 1 Introduction Probabilistic Latent Semantic Analysis (plsa) is a technique from the category of topic models. Its main goal is to model cooccurrence information

More information

Detecting Humans via Their Pose

Detecting Humans via Their Pose Detecting Humans via Their Pose Alessandro Bissacco Computer Science Department University of California, Los Angeles Los Angeles, CA 90095 bissacco@cs.ucla.edu Ming-Hsuan Yang Honda Research Institute

More information

Visual Object Detection

Visual Object Detection Visual Object Detection Ying Wu Electrical Engineering and Computer Science Northwestern University, Evanston, IL 60208 yingwu@northwestern.edu http://www.eecs.northwestern.edu/~yingwu 1 / 47 Visual Object

More information

SUPERVISED LEARNING: INTRODUCTION TO CLASSIFICATION

SUPERVISED LEARNING: INTRODUCTION TO CLASSIFICATION SUPERVISED LEARNING: INTRODUCTION TO CLASSIFICATION 1 Outline Basic terminology Features Training and validation Model selection Error and loss measures Statistical comparison Evaluation measures 2 Terminology

More information

Pedestrian Density Estimation by a Weighted Bag of Visual Words Model

Pedestrian Density Estimation by a Weighted Bag of Visual Words Model Pedestrian Density Estimation by a Weighted Bag of Visual Words Model Shilin Zhang and Xunyuan Zhang image representation termed bag of visual words integrating weighting scheme and spatial pyramid co-occurrence,

More information

The state of the art and beyond

The state of the art and beyond Feature Detectors and Descriptors The state of the art and beyond Local covariant detectors and descriptors have been successful in many applications Registration Stereo vision Motion estimation Matching

More information

IV Course Spring 14. Graduate Course. May 4th, Big Spatiotemporal Data Analytics & Visualization

IV Course Spring 14. Graduate Course. May 4th, Big Spatiotemporal Data Analytics & Visualization Spatiotemporal Data Visualization IV Course Spring 14 Graduate Course of UCAS May 4th, 2014 Outline What is spatiotemporal data? How to analyze spatiotemporal data? How to visualize spatiotemporal data?

More information

Unsupervised Activity Perception by Hierarchical Bayesian Models

Unsupervised Activity Perception by Hierarchical Bayesian Models Unsupervised Activity Perception by Hierarchical Bayesian Models Xiaogang Wang Xiaoxu Ma Eric Grimson Computer Science and Artificial Intelligence Lab Massachusetts Tnstitute of Technology, Cambridge,

More information

Lecture 8: Clustering & Mixture Models

Lecture 8: Clustering & Mixture Models Lecture 8: Clustering & Mixture Models C4B Machine Learning Hilary 2011 A. Zisserman K-means algorithm GMM and the EM algorithm plsa clustering K-means algorithm K-means algorithm Partition data into K

More information

Fisher Vector image representation

Fisher Vector image representation Fisher Vector image representation Machine Learning and Category Representation 2014-2015 Jakob Verbeek, January 9, 2015 Course website: http://lear.inrialpes.fr/~verbeek/mlcr.14.15 A brief recap on kernel

More information

Juergen Gall. Analyzing Human Behavior in Video Sequences

Juergen Gall. Analyzing Human Behavior in Video Sequences Juergen Gall Analyzing Human Behavior in Video Sequences 09. 10. 2 01 7 Juer gen Gall Instit u t e of Com puter Science III Com puter Vision Gr oup 2 Analyzing Human Behavior Analyzing Human Behavior Human

More information

Clustering with k-means and Gaussian mixture distributions

Clustering with k-means and Gaussian mixture distributions Clustering with k-means and Gaussian mixture distributions Machine Learning and Category Representation 2014-2015 Jakob Verbeek, ovember 21, 2014 Course website: http://lear.inrialpes.fr/~verbeek/mlcr.14.15

More information

Unsupervised Activity Perception in Crowded and Complicated Scenes Using Hierarchical Bayesian Models

Unsupervised Activity Perception in Crowded and Complicated Scenes Using Hierarchical Bayesian Models SUBMISSION TO IEEE TRANS. ON PATTERN ANALYSIS AND MACHINE INTELLIGENCE 1 Unsupervised Activity Perception in Crowded and Complicated Scenes Using Hierarchical Bayesian Models Xiaogang Wang, Xiaoxu Ma,

More information

9/12/17. Types of learning. Modeling data. Supervised learning: Classification. Supervised learning: Regression. Unsupervised learning: Clustering

9/12/17. Types of learning. Modeling data. Supervised learning: Classification. Supervised learning: Regression. Unsupervised learning: Clustering Types of learning Modeling data Supervised: we know input and targets Goal is to learn a model that, given input data, accurately predicts target data Unsupervised: we know the input only and want to make

More information

Face detection and recognition. Detection Recognition Sally

Face detection and recognition. Detection Recognition Sally Face detection and recognition Detection Recognition Sally Face detection & recognition Viola & Jones detector Available in open CV Face recognition Eigenfaces for face recognition Metric learning identification

More information

Anomaly (outlier) detection. Huiping Cao, Anomaly 1

Anomaly (outlier) detection. Huiping Cao, Anomaly 1 Anomaly (outlier) detection Huiping Cao, Anomaly 1 Outline General concepts What are outliers Types of outliers Causes of anomalies Challenges of outlier detection Outlier detection approaches Huiping

More information

10-810: Advanced Algorithms and Models for Computational Biology. Optimal leaf ordering and classification

10-810: Advanced Algorithms and Models for Computational Biology. Optimal leaf ordering and classification 10-810: Advanced Algorithms and Models for Computational Biology Optimal leaf ordering and classification Hierarchical clustering As we mentioned, its one of the most popular methods for clustering gene

More information

Optical Acceleration for Motion Description in Videos

Optical Acceleration for Motion Description in Videos Optical Acceleration for Motion Description in Videos Anitha Edison & Jiji C. V. College of Engineering, Trivandrum Kerala, India anithaedison@cet.ac.in, jijicv@cet.ac.in Abstract Modern techniques for

More information

Intro. ANN & Fuzzy Systems. Lecture 15. Pattern Classification (I): Statistical Formulation

Intro. ANN & Fuzzy Systems. Lecture 15. Pattern Classification (I): Statistical Formulation Lecture 15. Pattern Classification (I): Statistical Formulation Outline Statistical Pattern Recognition Maximum Posterior Probability (MAP) Classifier Maximum Likelihood (ML) Classifier K-Nearest Neighbor

More information

Leaving Some Stones Unturned: Dynamic Feature Prioritization for Activity Detection in Streaming Video

Leaving Some Stones Unturned: Dynamic Feature Prioritization for Activity Detection in Streaming Video In Proceedings of the European Conference on Computer Vision (ECCV), 26 Leaving Some Stones Unturned: Dynamic Feature Prioritization for Activity Detection in Streaming Video Yu-Chuan Su and Kristen Grauman

More information

Activity Recognition using Dynamic Subspace Angles

Activity Recognition using Dynamic Subspace Angles Activity Recognition using Dynamic Subspace Angles Binlong Li, Mustafa Ayazoglu, Teresa Mao, Octavia I. Camps and Mario Sznaier Dept. of Electrical and Computer Engineering, Northeastern University, Boston,

More information

Introduction to Machine Learning

Introduction to Machine Learning Introduction to Machine Learning CS4731 Dr. Mihail Fall 2017 Slide content based on books by Bishop and Barber. https://www.microsoft.com/en-us/research/people/cmbishop/ http://web4.cs.ucl.ac.uk/staff/d.barber/pmwiki/pmwiki.php?n=brml.homepage

More information

A Tutorial on Learning with Bayesian Networks

A Tutorial on Learning with Bayesian Networks A utorial on Learning with Bayesian Networks David Heckerman Presented by: Krishna V Chengavalli April 21 2003 Outline Introduction Different Approaches Bayesian Networks Learning Probabilities and Structure

More information

Graphical Object Models for Detection and Tracking

Graphical Object Models for Detection and Tracking Graphical Object Models for Detection and Tracking (ls@cs.brown.edu) Department of Computer Science Brown University Joined work with: -Ying Zhu, Siemens Corporate Research, Princeton, NJ -DorinComaniciu,

More information

SYSTEMATIC CONSTRUCTION OF ANOMALY DETECTION BENCHMARKS FROM REAL DATA. Outlier Detection And Description Workshop 2013

SYSTEMATIC CONSTRUCTION OF ANOMALY DETECTION BENCHMARKS FROM REAL DATA. Outlier Detection And Description Workshop 2013 SYSTEMATIC CONSTRUCTION OF ANOMALY DETECTION BENCHMARKS FROM REAL DATA Outlier Detection And Description Workshop 2013 Authors Andrew Emmott emmott@eecs.oregonstate.edu Thomas Dietterich tgd@eecs.oregonstate.edu

More information

Clustering with k-means and Gaussian mixture distributions

Clustering with k-means and Gaussian mixture distributions Clustering with k-means and Gaussian mixture distributions Machine Learning and Object Recognition 2017-2018 Jakob Verbeek Clustering Finding a group structure in the data Data in one cluster similar to

More information

Template-Based Representations. Sargur Srihari

Template-Based Representations. Sargur Srihari Template-Based Representations Sargur srihari@cedar.buffalo.edu 1 Topics Variable-based vs Template-based Temporal Models Basic Assumptions Dynamic Bayesian Networks Hidden Markov Models Linear Dynamical

More information

Riemannian Metric Learning for Symmetric Positive Definite Matrices

Riemannian Metric Learning for Symmetric Positive Definite Matrices CMSC 88J: Linear Subspaces and Manifolds for Computer Vision and Machine Learning Riemannian Metric Learning for Symmetric Positive Definite Matrices Raviteja Vemulapalli Guide: Professor David W. Jacobs

More information

Randomized Algorithms

Randomized Algorithms Randomized Algorithms Saniv Kumar, Google Research, NY EECS-6898, Columbia University - Fall, 010 Saniv Kumar 9/13/010 EECS6898 Large Scale Machine Learning 1 Curse of Dimensionality Gaussian Mixture Models

More information

Bayesian Networks Inference with Probabilistic Graphical Models

Bayesian Networks Inference with Probabilistic Graphical Models 4190.408 2016-Spring Bayesian Networks Inference with Probabilistic Graphical Models Byoung-Tak Zhang intelligence Lab Seoul National University 4190.408 Artificial (2016-Spring) 1 Machine Learning? Learning

More information

Pattern recognition. "To understand is to perceive patterns" Sir Isaiah Berlin, Russian philosopher

Pattern recognition. To understand is to perceive patterns Sir Isaiah Berlin, Russian philosopher Pattern recognition "To understand is to perceive patterns" Sir Isaiah Berlin, Russian philosopher The more relevant patterns at your disposal, the better your decisions will be. This is hopeful news to

More information

Statistical Methods for NLP

Statistical Methods for NLP Statistical Methods for NLP Sequence Models Joakim Nivre Uppsala University Department of Linguistics and Philology joakim.nivre@lingfil.uu.se Statistical Methods for NLP 1(21) Introduction Structured

More information

Boosting & Deep Learning

Boosting & Deep Learning Boosting & Deep Learning Ensemble Learning n So far learning methods that learn a single hypothesis, chosen form a hypothesis space that is used to make predictions n Ensemble learning à select a collection

More information

L11: Pattern recognition principles

L11: Pattern recognition principles L11: Pattern recognition principles Bayesian decision theory Statistical classifiers Dimensionality reduction Clustering This lecture is partly based on [Huang, Acero and Hon, 2001, ch. 4] Introduction

More information

Kernel Density Topic Models: Visual Topics Without Visual Words

Kernel Density Topic Models: Visual Topics Without Visual Words Kernel Density Topic Models: Visual Topics Without Visual Words Konstantinos Rematas K.U. Leuven ESAT-iMinds krematas@esat.kuleuven.be Mario Fritz Max Planck Institute for Informatics mfrtiz@mpi-inf.mpg.de

More information

Image Processing 1 (IP1) Bildverarbeitung 1

Image Processing 1 (IP1) Bildverarbeitung 1 MIN-Fakultät Fachbereich Informatik Arbeitsbereich SAV/BV KOGS Image Processing 1 IP1 Bildverarbeitung 1 Lecture : Object Recognition Winter Semester 015/16 Slides: Prof. Bernd Neumann Slightly revised

More information

Object Detection Grammars

Object Detection Grammars Object Detection Grammars Pedro F. Felzenszwalb and David McAllester February 11, 2010 1 Introduction We formulate a general grammar model motivated by the problem of object detection in computer vision.

More information

Recent advances in Time Series Classification

Recent advances in Time Series Classification Distance Shapelet BoW Kernels CCL Recent advances in Time Series Classification Simon Malinowski, LinkMedia Research Team Classification day #3 S. Malinowski Time Series Classification 21/06/17 1 / 55

More information

Latent Geographic Feature Extraction from Social Media

Latent Geographic Feature Extraction from Social Media Latent Geographic Feature Extraction from Social Media Christian Sengstock* Michael Gertz Database Systems Research Group Heidelberg University, Germany November 8, 2012 Social Media is a huge and increasing

More information

20 Unsupervised Learning and Principal Components Analysis (PCA)

20 Unsupervised Learning and Principal Components Analysis (PCA) 116 Jonathan Richard Shewchuk 20 Unsupervised Learning and Principal Components Analysis (PCA) UNSUPERVISED LEARNING We have sample points, but no labels! No classes, no y-values, nothing to predict. Goal:

More information

Action Attribute Detection from Sports Videos with Contextual Constraints

Action Attribute Detection from Sports Videos with Contextual Constraints YU, ET AL: ACTION ATTRIBUTE DETECTION FROM SPORTS VIDEOS 1 Action Attribute Detection from Sports Videos with Contextual Constraints Xiaodong Yu 1 xiaodong_yu@cable.comcast.com Ching Lik Teo 2 cteo@cs.umd.edu

More information

Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials

Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials Efficient Inference in Fully Connected CRFs with Gaussian Edge Potentials Philipp Krähenbühl and Vladlen Koltun Stanford University Presenter: Yuan-Ting Hu 1 Conditional Random Field (CRF) E x I = φ u

More information

Natural Language Processing

Natural Language Processing Natural Language Processing Word vectors Many slides borrowed from Richard Socher and Chris Manning Lecture plan Word representations Word vectors (embeddings) skip-gram algorithm Relation to matrix factorization

More information

Non-parametric Methods

Non-parametric Methods Non-parametric Methods Machine Learning Alireza Ghane Non-Parametric Methods Alireza Ghane / Torsten Möller 1 Outline Machine Learning: What, Why, and How? Curve Fitting: (e.g.) Regression and Model Selection

More information

Information Extraction from Text

Information Extraction from Text Information Extraction from Text Jing Jiang Chapter 2 from Mining Text Data (2012) Presented by Andrew Landgraf, September 13, 2013 1 What is Information Extraction? Goal is to discover structured information

More information

Predicting New Search-Query Cluster Volume

Predicting New Search-Query Cluster Volume Predicting New Search-Query Cluster Volume Jacob Sisk, Cory Barr December 14, 2007 1 Problem Statement Search engines allow people to find information important to them, and search engine companies derive

More information

DISCRIMINATIVE DECORELATION FOR CLUSTERING AND CLASSIFICATION

DISCRIMINATIVE DECORELATION FOR CLUSTERING AND CLASSIFICATION DISCRIMINATIVE DECORELATION FOR CLUSTERING AND CLASSIFICATION ECCV 12 Bharath Hariharan, Jitandra Malik, and Deva Ramanan MOTIVATION State-of-the-art Object Detection HOG Linear SVM Dalal&Triggs Histograms

More information

Handling imprecise and uncertain class labels in classification and clustering

Handling imprecise and uncertain class labels in classification and clustering Handling imprecise and uncertain class labels in classification and clustering Thierry Denœux 1 1 Université de Technologie de Compiègne HEUDIASYC (UMR CNRS 6599) COST Action IC 0702 Working group C, Mallorca,

More information

lecture 6: modeling sequences (final part)

lecture 6: modeling sequences (final part) Natural Language Processing 1 lecture 6: modeling sequences (final part) Ivan Titov Institute for Logic, Language and Computation Outline After a recap: } Few more words about unsupervised estimation of

More information

Statistical learning. Chapter 20, Sections 1 4 1

Statistical learning. Chapter 20, Sections 1 4 1 Statistical learning Chapter 20, Sections 1 4 Chapter 20, Sections 1 4 1 Outline Bayesian learning Maximum a posteriori and maximum likelihood learning Bayes net learning ML parameter learning with complete

More information

CS534 Machine Learning - Spring Final Exam

CS534 Machine Learning - Spring Final Exam CS534 Machine Learning - Spring 2013 Final Exam Name: You have 110 minutes. There are 6 questions (8 pages including cover page). If you get stuck on one question, move on to others and come back to the

More information

COMP90051 Statistical Machine Learning

COMP90051 Statistical Machine Learning COMP90051 Statistical Machine Learning Semester 2, 2017 Lecturer: Trevor Cohn 24. Hidden Markov Models & message passing Looking back Representation of joint distributions Conditional/marginal independence

More information

Hidden Markov Models Part 1: Introduction

Hidden Markov Models Part 1: Introduction Hidden Markov Models Part 1: Introduction CSE 6363 Machine Learning Vassilis Athitsos Computer Science and Engineering Department University of Texas at Arlington 1 Modeling Sequential Data Suppose that

More information

A Comparative Evaluation of Anomaly Detection Techniques for Sequence Data. Technical Report

A Comparative Evaluation of Anomaly Detection Techniques for Sequence Data. Technical Report A Comparative Evaluation of Anomaly Detection Techniques for Sequence Data Technical Report Department of Computer Science and Engineering University of Minnesota 4-192 EECS Building 200 Union Street SE

More information

Maximally Stable Local Description for Scale Selection

Maximally Stable Local Description for Scale Selection Maximally Stable Local Description for Scale Selection Gyuri Dorkó and Cordelia Schmid INRIA Rhône-Alpes, 655 Avenue de l Europe, 38334 Montbonnot, France {gyuri.dorko,cordelia.schmid}@inrialpes.fr Abstract.

More information

Text Mining. Dr. Yanjun Li. Associate Professor. Department of Computer and Information Sciences Fordham University

Text Mining. Dr. Yanjun Li. Associate Professor. Department of Computer and Information Sciences Fordham University Text Mining Dr. Yanjun Li Associate Professor Department of Computer and Information Sciences Fordham University Outline Introduction: Data Mining Part One: Text Mining Part Two: Preprocessing Text Data

More information

Machine Learning - MT & 14. PCA and MDS

Machine Learning - MT & 14. PCA and MDS Machine Learning - MT 2016 13 & 14. PCA and MDS Varun Kanade University of Oxford November 21 & 23, 2016 Announcements Sheet 4 due this Friday by noon Practical 3 this week (continue next week if necessary)

More information