Julián Ramos Cózar

A Clustering Technique for Video Copy Detection

In this work, a new method for detecting copies of a query video in a videos database is proposed. It includes a new clustering technique that groups frames with similar visual content, maintaining their temporal order. Applying this technique, a keyframe is extracted for each cluster of the query video. Keyframe choice is carried out by selecting the frame in the cluster with maximum similarity to the rest of frames in the cluster. Then, keyframes are compared to target videos frames in order to extract similarity regions in the target video. Relaxed temporal constraints are subsequently applied to the calculated regions in order to identify the copy sequence. The reliability and performance of the method has been tested by using several videos from the MPEG-7 Content Set, encoded with different frame sizes, bit rates and frame rates. Results show that our method obtains a significant improvement with respect to previous approaches in both achieved precision and computation time.

Logotype detection to support semantic-based video annotation

In conventional video production, logotypes are used to convey information about content originator or the actual video content. Logotypes contain information that is critical to infer genre, class and other important semantic features of video. This paper presents a framework to support semantic-based video classification and annotation. The backbone of the proposed framework is a technique for logotype extraction and recognition. The method consists of two main processing stages. The first stage performs temporal and spatial segmentation by calculating the minimal luminance variance region (MVLR) for a set of frames. Non-linear diffusion filters (NLDF) are used at this stage to reduce noise in the shape of the logotype. In the second stage, logotype classification and recognition are achieved. The earth movers distance (EMD) is used as a metric to decide if the detected MLVR belongs to one of the following logotype categories: learned or candidate. Learned logos are semantically annotated shapes available in the database. The semantic characterization of such logos is obtained through an iterative learning process. Candidate logos are non-annotated shapes extracted during the first processing stage. They are assigned to clusters grouping different instances of logos of similar shape. Using these clusters, false logotypes are removed and different instances of the same logo are averaged to obtain a unique prototype representing the underlying noisy cluster. Experiments involving several hours of MPEG video and around 1000 of candidate logotypes have been carried out in order to show the robustness of both detection and classification processes.

Planar 3D object detection by using the generalized Hough transform

In this work we propose a new method to detect arbitrary planar shapes from a previous template and calculate the parameters that define the transformations between the new image and the template. The image contains a perspective projection of the template subjected to two angle transformations, called tilt and pan, a displacement, a rotation and a scaling. The method uncouples parameter calculation to improve computational requirements by comparing invariant information from the template and the image. The generalized Hough transform is used to compare this information and to vote into a parameter space.

A TV-logo classification and learning system

Logotypes superimposed to broadcasted videos supply important information for semantic video annotation, such as the content creator. In this work a novel logo classification and learning system for TV broadcast videos is presented. Logos are segmented from the video stream but scale change, position shift, clutter and noise makes difficult to classify and to recognize them. Several robust features that use edges and shape information have been selected, and a Bayesian network classifier is used to classify the logos. New logos are recognized as such for the first time they appear and passed to a semi-supervised learning system. The learning process clusters the set of new logos to group different instances of the same new logo. A logo model is obtained for each cluster that must be validated by a human to incorporate them into the classification system. Comprehensive tests with a set of 724 TV logos show the high performance of our classification and learning system.

Video Cataloging Based on Robust Logotype Detection

In this paper a technique for video cataloging based on logo detection is shown. No a priori knowledge about shape or spatial-temporal location of logos is assumed. The method implements a new algorithm for online logo detection based on temporal and spatial segmentation of broadcasted videos. Temporal segmentation identifies constant luminance regions within video frames while spatial segmentation helps to refine previous segmented regions. In a final step, identified logos are searched in a database and classified into candidate or learnt logotypes. Learnt logos can be directly tracked through the video. Candidate logotypes are assigned to a cluster of similar logos. After a promotion process, all the candidate logos belonging to the same cluster are used to create a new learnt logotype.

Detection of arbitrary planar shapes with 3D pose

Abstract In this work, two new methods to detect objects under perspective and scaled orthographic projection are shown. They also calculate the parameters of the transformations the object has undergone. The methods are based on the use of the Generalized Hough Transform (GHT) that compares a template with a projected image. The computational requirements of the algorithms are reduced by restricting the transformation to the image edge points and using invariant information during the comparison process. Moreover, a multipass design of the algorithms speeds-up the parameter calculations.

Visual words selection for human action classification

Human action classification is an important task in computer vision. The Bag-of-Words model uses spatio-temporal features assigned to visual words of a vocabulary and some classification algorithm to attain this goal. In this work we have studied the effect of reducing the vocabulary size using a video word ranking method. We have applied this method to the KTH dataset to obtain a vocabulary with more descriptive words where the representation is more compact and efficient. Two feature descriptors, STIP and MoSIFT, and two classifiers, KNN and SVM, have been used to check the validity of our approach. Results for different vocabulary sizes show an improvement of the recognition rate whilst reducing the number of words as non-descriptive words are removed. Additionally, state-of-the-art performances are reached with this new compact vocabulary representation.

Planar Object Detection under Scaled Orthographic Projection

In this work a new method to detect objects under scaled orthographic projections is shown. It also calculates the parameters of the transformations the object has suffered. The method is based on the use of the Generalized Hough Transform (GHT) that compares a template with a projected image. The computational requirements of the algorithm are reduced by restricting the transformation to the template edge points and using invariant information during the comparison process. This information is obtained from a precomputed table of the template that is directly transformed and compared with the image table. Moreover, a multiresolution design of the algorithm speeds-up the parameters calculation.

Low-textured regions detection for improving stereoscopy algorithms

The main goal of stereoscopy algorithms is the calculation of the disparity map between two frames corresponding to the same scene, and captured simultaneously by two different cameras. The different position (disparity) where common scene points are projected in both camera sensors can be used to calculate the depth of the scene point. Many algorithms calculate the disparity of corresponding points in both frames relying on the existence of similar textured areas around the pixels to be analyzed. Unfortunately, real images present large areas with low texture, which hinder the calculation of the disparity map. In this paper we present a method that employs a set of local textures to build a classifier that is able to select reliable pixels where the disparity can be accurately calculated, improving the precision of the scene map obtained by the stereoscopic technique.

Kernel-based object tracking using a simple fuzzy color histogram

In this paper, we present an approach for kernel-based object tracking using the HSV color space as the feature space and fuzzy color histograms as feature vectors. These histograms are more robust to illumination changes and quantization errors than common histograms. To avoid a significant increase in the computational complexity, a simple fuzzy membership function is used. The efficiency of this approach is demonstrated using videos from the PETS database and comparing the results using the fuzzy color histogram and the common color histogram.