Another List for Advanced Topics in Computer Vision in 2010

We provide an overview of the current state-of-the-art of computer vision by studying a set of cutting-edge advanced computer vision topics. Nine selected computer vision research topics reflect the current state in machine vision. The main objective is to review cutting-edge computer vision research through lectures covering the underlying mathematical concepts and representative algorithms, paper reading, and implementation.

 

Semi-supervised Learning

Sparsity Induced Similarity Measure for Label Propagation

Unsupervised Improvements of Visual Detectors using Co-training

Robust Multi-Class Transductive Learning with Graphs

 

Compressive Sensing

A Compressive Sensing Approach for Expression-Invariant Face Recognition

Discriminative Learned Dictionaries for Local Image Analysis

Compressive Sensing for Background Subtraction

Image Super-Resolution as Sparse Representation of Raw Image Patches

Implicit elastic matching with random projections for pose-variant face recognition

 

Segmentation

Increased Discrimination in Level Set Methods with Embedded Conditional Random Fields

3D Pose Estimation and Segmentation using Specular Cues

An efficient algorithm for Co-segmentation

Image Segmentation with A Bounding Box Prior

Spectral clustering of linear subspaces for motion segmentation

 

Optical Flow

Large Displacement Optical Flow

Contextual Flow

Local Grouping for Optical Flow

 

Human Detection

Human Detection Using Partial Least Squares Analysis

Automatic Annotation of Human Actions in Video

An HOG-LBP Human Detector with Partial Occlusion Handling

Robust Tracking-by-Detection using a Detector Confidence Particle Filter

 

Registration, Matching, and Alignment

Image Registration by Minimization of Residual Complexity

Simultaneous Alignment and Clustering for an Image Ensemble

Canonical TimeWarping for Alignment of Human Behavior

Temporal Segmentation and Activity Classification from First-person Sensing

 

Wavelet Based Topics

Wavelet Energy Map: A Robust Support for Multi-modal Registration of Medical Images

An Efficient Algorithm for Compressed MR Imaging using Total Variation and Wavelet

Modeling the Marginal Distributions of Complex Wavelet Coefficient Magnitudes for the Classification of Zoom-Endoscopy Images

Hierarchical 3D Diffusion Wavelets Shape Priors

 

Multi-target Tracking

You'll NeverWalk Alone: Modeling Social Behavior for Multi-target Tracking

Coupled Detection and Trajectory Estimation for Multi-Object Tracking

Learning to Associate: HybridBoosted Multi-Target Tracker for Crowded Scene

 

Patterns

Sparse Subspace Clustering

Kernel Methods for Weakly Supervised Mean Shift Clustering

In Defense of Nearest-Neighbor Based Image Classification

Mode-Detection via Median-Shift

A Tensor-Based Algorithm for High-Order Graph Matching

Higher-Order Clique Reduction in Binary Graph Cut

Beyond Pairwise Energies: Efficient Optimization for Higher-order MRFs

 

Recognition

Distinctive Image Features from Scale-Invariant Keypoints
Object Class Recognition by Unsupervised Scale-Invariant Learning
Video Google: a Text Retrieval Approach to Object Matching in Videos
Learning Object Categories from Google's Image Search
Adapted Vocabularies for Generic Visual Categorization
Discovering Objects and Their Location in Images
Object Retrieval with Large Vocabularies and Fast Spatial Matching
Scalable Recognition with a Vocabulary Tree ( Relevant Reference)
Beyond Bag of Features: Spatial Pyramid Matching for Recognizing Natural Scene Categories
Multiple Object Class Detection with a Generative Model
Shape Matching and Object Recognition using Low Distortion Correspondences
Contour-based learning for object detection
A boundary-Fragment-Model for Object Detection
Beyond Local Appearance: Category Recognition from Pairwise Interactions of Simple Features

 

From

[1] http://ranger.uta.edu/~heng/CSE6369.html

[2] http://www.cise.ufl.edu/class/cis6930fa07atc/