Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade...

24
Low-Rank Based Algorithms for Rectification, Repetition Detection and De-Noising in Urban Images A dissertation proposal by Juan Liu

Transcript of Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade...

Page 1: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Low-Rank Based Algorithms for Rectification, Repetition Detection and De-Noising in Urban Images

A dissertation proposal by

Juan Liu

Page 2: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Committee

Committee:

Professor Ioannis Stamos, Mentor, Hunter College

Professor Yingli Tian, City College

Professor Zhigang Zhu, City College

Outside Member:

Professor Emmanouil Z. Psarakis, University of Patras, Greece

Page 3: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Current Work

2D image rectification

Façade texture selection

Efficient Kronecker Product model

Automatic repeated patterns detection

Reconstruction & photorealistic rendering of urban environments, symmetry detection, hole filling, etc.

Page 4: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Façade Image Rectification

Vanishing points (VPs) detection

TILT

Vanishing points Draw m hypothetical lines at angular intervals for each vanishing point

Output

Page 5: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Façade Texture Selection

Let C be the non-negative m×m matrix, with each element representing the number of Harris corners inside a block

Slide a window of size r×c along C

Compute the sample mean deviation of the sample median of matrix C, μC

Page 6: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Façade Texture Selection

Page 7: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Kronecker Product Model

Page 8: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

A Toy Example

Input facade Repeated patterns

Page 9: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Repeated Patterns Detection via Kronecker Product Model

• Define the cost function:

• The minimization problem:

Page 10: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Solution to The Minimization Problem

Theorem 1: Let be the Singular Value Decomposition of the rearranged counterpart of matrix . Then partition matrices , patterns and weighting factors should satisfy the following:

Page 11: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Solution to The Minimization Problem

Lemma 1: Assuming that matrices are known, then the optimal and are related as follows:

Page 12: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Spatial Periods Estimation via Cross Correlation

• Consider that are known, then we can estimate the periods for facade partition

3×10 partition blocks Fi,j

Column-wise Cross-Correlation sequences: distance between the

adjacent peaks provides the period information.

Partition

Page 13: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Block Vectorization

• Vectorization of partition blocks

• The equivalent cost function:

Vectorization

rank = 30

Page 14: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Estimating K by Clustering

Reshape

rank = 4

Page 15: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Estimation of Mκ, κ =1, 2, …, K★

• Reshape the K★ indicators vectors from Algorithm 2

Page 16: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Computing Patterns and Weighting Factors

• Using Lemma 1 and Algorithm 2

Group 1

Group 2

Group 3

Group 4

Classification

Classification

Classification

Classification

Refinement

Refinement

Refinement

Refinement

1-0 patterns re-construction via Kronecker Product Model

Page 17: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Experiments and Evaluation

• Experiment 1: image rectification and texture selection

• Experiment 2: repeated patterns detection 89 façade images with ground-truth

Success rate 96%

Pixel-wise comparison

Success threshold: 91% match with ground-truth

Page 18: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Results

Page 19: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Proposed Work

Improve the estimation of K★

Extend the method to model nested patterns

Apply the model to 3D point clouds

Page 20: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Improve the estimation of K★

• Limitation: the accuracy and efficiency is drawn back by

– K-means clustering algorithm

Page 21: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Period Computation for Nested Patterns

• The current model may cut a bigger pattern into pieces due to that:– The most frequently appeared pattern dominates the final period

Page 22: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Apply the model to 3D point clouds

Page 23: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Timeline

• My work plan to complete the dissertation is arranged as follows:– Jan. 9 2015 - Jan. 20 2015:

• Finalize the implementation of new methods.

• Run experiments to test the new methods.

• Analyze the performance based upon experiment results.

– Jan. 21 2015 - Feb. 21 2015: Complete the rest part of my thesis and prepare for defense.

– Finally: Defend in March 2015.

Page 24: Low-Rank Based Algorithms for Rectification, Repetition ...€¦ · 2D image rectification Façade texture selection Efficient Kronecker Product model Automatic repeated patterns

Related Publications

• One paper “Automatic Kronecker product model based detection of repeated patterns in 2D urban images” that is related to this proposal has been published in IEEE International Conference on Computer Vision (ICCV) 2013 (accepatance rate 28%, 1600 submissions).

• Another related paper has been submitted to the IEEE Transactions on Pattern Analysis and Machine Intelligence (TPAMI) recently.