Local Feature Learning

Local Feature Learning

Local Feature Learning Gustavo Carneiro Tutorial ICIP 2013 Local Feature Learning and Non-rigid Matching Why Local Features? Model Test 1 Test 2 GloBal Local MORE ROBUST MORE DISTINCTIVE 2-dim feature space G 2-dim feature space L G. Carneiro - University of Adelaide 2 Local Feature • Limited spaal support + RoBust to changes and parPal occlusion - Discriminang power (compensated By # descriptors) • Applicaons – Visual classificaon – Image matching G. Carneiro - University of Adelaide 3 Representaon • An image is represented By a set of N descriptors (or parts) G. Carneiro - University of Adelaide 4 Representaon • An image is represented By a set of N descriptors (or parts) f1=[ ,x1,θ1,σ1] scale appearance geometry dominant orientation position 5 Representaon • An image is represented By a set of N descriptors (or parts) f2 = [ a2 , g2 ] G. Carneiro - University of Adelaide 6 Representaon • An image is represented By a set of N descriptors (or parts) F={ f1 , f2 , f3 , ... , fN } G. Carneiro - University of Adelaide 7 Image Matching [Schaffalitzky & Zisserman, ECCV02] G. Carneiro - University of Adelaide 8 Visual Classificaon • Instance-Based recogniPon [Lowe,IJCV04] • Class recogniPon G. Carneiro - University of Adelaide 9 Objecve Funcons • Image matching – Maximize precision and recall of feature matching • Instance-Based recogniPon and class recogniPon – Minimize classificaon error G. Carneiro - University of Adelaide 10 Hand-designed Local Features 90’s to 2000’s • ‘Where’ step – Repeatable Harris “corners” [Harris88] DoG [Lowe99,04] Sum square diffs Diff of Gaussians RoBust to rotaon RoBust to scale • ‘What’ step – RoBust – Disncve SIFT [Lowe99,04] 128 dimensions RoBust to rigid transforms and Brightness 11 Quesons • Are these hand-designed features opPmal? • Image matching and instance-Based recogniPon (Does it maximize matching precision and recall?) – If disPncPve & roBust then max. precision and recall • Class recogniPon (Does it minimize classificaon error?) – If roBust & disPncPve then min. classificaon error G. Carneiro - University of Adelaide 12 PosiPve Evidence viewpoint scale [Mikolajczyk and Schmid, PAMI’05] • SIFT-like features showed superior performance – Dominates matching and classificaon applicaons G. Carneiro - University of Adelaide 13 Matching ProBlem • Building Rome in a Day [Agarwal et al. ICCV09] – Reconstruct 3D scenes from large collecPon of images • Matching Based on SIFT features • Image similarity also Based on SIFT features G. Carneiro - University of Adelaide 14 Class RecogniPon • Bag of Features [Sivic and Zisserman, ICCV03] SVM Classifier • Also use SIFT features G. Carneiro - University of Adelaide 15 However • Good features to track [Shi and Tomasi,CVPR94] – Feature selecPon Based on model similarity • Detectability, Uniqueness, and Reliability [OBah and Ikeuchi,PAMI97] 16 But these are not SIFT… • True, But D. Lowe noPced something similar about the discriminang power of SIFT • Not all descriptors have the same discriminang power. • Can a similar thing Be said about the roBustness properPes of the feature? G. Carneiro - University of Adelaide 17 Explicit Characterizaon of RoBustness and Discriminang Power [Carneiro and Jepson, CVPR05] • RoBustness: Pon(sf(fl,fo);fl) ˜ Pβ(sf(fl,fo);aon,bon) • Disncveness : Poff(sf(fl,fo);fl) ˜ Pβ(sf(fl,fo);aoff,boff) • Detectability: Pdet(xl) P P Feature vector #260 on off Pdet(xl)=87% Feature vector #540 Pdet(xl)=67% 18 Phase correlaon Train Classifier to Select and Characterize “Good” Features • P(OBj|Match,Img) =(1/Z)P(Match|OBj,I)P(OBj|Img) 19 SelecPng and Characterizing Good Features… • Does it lead to more effecPve matching? • Does it lead to more effecPve classificaon? • Why can’t we learn the features By maximizing the actual oBjecPve funcPon? – Instead of designing and charactering individual features G. Carneiro - University of Adelaide 20 In the Beginning… • Perceptron [RosenBla 57] fi (x) wi y = sign(! fi (x)wi + b) • Features to use? – Again, hand-designed… 21 Learning Input Features? • Perceptron can only deal with linear proBlems • MulP-layer perceptron (non-linear acPvaon funcPons)? – Can deal with more complex proBlems! – Can we finally learn the input features from the image? (1) wi (2) wi ` … ! (!viwi + b) 22 Back-propagaon [Rumelhart et al. 86] • Algorithm that allowed training of mulP-layer perceptron could not handle more than 1-2 hidden layers – Long Pme to converge (if it converges at all) • Back to hand-designing/selecPon/ characterizaon of features… G. Carneiro - University of Adelaide 23 TradiPonal Methods • Matching ProBlems Hand-designed Features Outlier Matching RejecPon Hand-designed Features • Visual Classificaon ProBlems Hand-designed Supervised Features Classifier G. Carneiro - University of Adelaide 24 Race is on for the “Best” Hand-designed Features • No transformaon – Gray values • Frequency domain – Discrete Fourier transform (DFT), Discrete Cosine Transform (DCT), Discrete Wavelet Transform (DWT) • Good ReconstrucPon and Uncorrelaon – Principal Components Analysis (PCA) • Good ReconstrucPon and Independence – Independent Component Analysis (ICA), sparse coding • Linear class separability – Linear Discriminant Analysis (LDA) • Gradient Orientaon Histogram – SIFT, HOG, GLOH, RIFT, etc. • Image DifferenPals – Local jets • Sampling and representaon variaons – RoBust and disPncPve G. Carneiro - University of Adelaide 25 Also for the “Best” Classifiers • Generave Classifiers – Mixture Model – Naïve Bayes • Discriminave Classifiers – LogisPc Regression – MulP-layer Perceptron – Nearest NeighBor – Support Vector Machine – Boosng – Random Forest G. Carneiro - University of Adelaide 26 Outlier RejecPon • Random Sample Consensus (RANSAC) [Fischler and Bolles, 81; Torr and Murray, IJCV’97] • MSAC (M-esPmator) , MLESAC (max likelihood), IMPSAC (importance sampling), etc. • Consequence of feature matching failures • More about this later on Dr. Chin’s session G. Carneiro - University of Adelaide 27 Feature Learning • Using the tradiPonal architecture, can we learn the features with the opPmizaon funcPon used for the classifier? • Matching – maximize feature roBustness and discriminang power • Classificaon – minimize classificaon error G. Carneiro - University of Adelaide 28 Feature Learning - Matching • AcPve Shape and Appearance Models [Cootes et al. 95,98] posion x Appearance g • ComBine shape and gray level in a single PCA space • Gradient descent to perform matching G. Carneiro - University of Adelaide 29 Feature Learning - Matching • FERNS [M. Özuysal et al. CVPR’07] • Semi-naïve Bayes classifier: G. Carneiro - University of Adelaide 30 D. Capel. 2009 Feature Learning - Matching • Explicitly learn a feature transform that is – RoBust and discriminang • Photo Tourism dataset [Snavely et al. SIGGRAPH’06] used by Winder et al. [Winder and Brown, CVPR’07,09 and PAMI’11] – More than 100,000 patches (3 scenes) – BackprojecPng 3D points to 2D images from scene reconstrucPons – Variaons in scene locaon, Brightness and parPal occlusion G. Carneiro - University of Adelaide 31 Feature Learning - Matching • Discriminave Learning of Local Features [Brown, Hua and Winder’PAMI11] Learning carried out To maximize AUC in ROC graph T-Block: Steerable filters, E-Block: Linear Distance metric learning Gradients, DoG, etc. N-Block: Normalizaon to account for photometric variaons S-Block: 32 (Linear) Distance Metric Learning [Chopra et al.CVPR05,GoldBerger et al.NIPS04, WeinBerger & Saul JMLR09] • Image patches: • Linear transform: • Distance in T space: G. Carneiro - University of Adelaide 33 Results [Winder and Brown’PAMI11] • Errors at 95% (% of error when 95% of TP are found) • In parenthesis: dimensionality G. Carneiro - University of Adelaide 34 Feature Learning - Matching • Learn feature transforms from non-linear distance metric learning [Carneiro,CVPR’10] – Uses original image input (no sampling stage, and focus on emBedding stage) – Use photo tourism dataset – Non-linear distance metric learning [Sugiyama JMLR07] G. Carneiro - University of Adelaide 35 ComBining Feature Spaces • Breiman’s idea about ensemBle classifiers [Breiman 01]: – comBine low-Bias, high-variance (unstable) classifiers to produce low-Bias, low-variance classifiers. • Distance G. Carneiro - University of Adelaide 36 IntuiPon Unkown target proBlem Small dist. Large dist. T G. Carneiro - University of Adelaide 37 Random training proBlem 1 IntuiPon Unkown target proBlem Small dist. Largedist. T G. Carneiro - University of Adelaide 38 Random training proBlem Experiments • Using cross validaon, – 50 training classes for training each feature space – 50 training feature spaces G. Carneiro - University of Adelaide 39 Experiments • Matching database of Mikolajczyk and Schmid 40 Feature Learning - Matching • Convexify Brown et al.’s opPmizaon funcPon [Simonyan et al. ECCV’12] • Use BoosPng to produce non-linear feature transform [Trzcinski et al. NIPS’12] • More to come J, But idea is the same – Given classes of local descriptors, find transformaon that keep features from the same class together, and separate features from disPnct classes G. Carneiro - University of Adelaide 41 Back to the Classificaon ProBlem • Feature selecPon to minimize classificaon error – RoBust Real-Pme OBject DetecPon [Viola and Jones, IJCV’01] • Feature extracPon to minimize Bayes error (BE) – Minimum BE facilitates training [Carneiro and Vasconcelos, CRV’05] • Feature Learning – Supervised ConvoluPonal Networks [Lecun 90s unPl today] G. Carneiro - University of Adelaide 42 Supervised Convolutional Network [Lecun, 90s until today]

View Full Text

Details

  • File Type
    pdf
  • Upload Time
    -
  • Content Languages
    English
  • Upload User
    Anonymous/Not logged-in
  • File Pages
    52 Page
  • File Size
    -

Download

Channel Download Status
Express Download Enable

Copyright

We respect the copyrights and intellectual property rights of all users. All uploaded documents are either original works of the uploader or authorized works of the rightful owners.

  • Not to be reproduced or distributed without explicit permission.
  • Not used for commercial purposes outside of approved use cases.
  • Not used to infringe on the rights of the original creators.
  • If you believe any content infringes your copyright, please contact us immediately.

Support

For help with questions, suggestions, or problems, please contact us