Virgínia Fernandes Mota

A tensor motion descriptor based on histograms of gradients and optical flow

This paper presents a new tensor motion descriptor only using optical flow and HOG3D information: no interest points are extracted and it is not based on a visual dictionary. We propose a new aggregation technique based on tensors. This is a double aggregation of tensor descriptors. The first one represents motion by using polynomial coefficients which approximates the optical flow. The other represents the accumulated data of all histograms of gradients of the video. The descriptor is evaluated by a classification of KTH, UCF11 and Hollywood2 datasets, using a SVM classifier. Our method reaches 93.2% of recognition rate with KTH, comparable to the best local approaches. For the UCF11 and Hollywood2 datasets, our recognition achieves fairly competitive results compared to local and learning based approaches.

A Tensor Based on Optical Flow for Global Description of Motion in Videos

Motion is one of the main characteristics that describe the semantic information of videos. In this work, a global video descriptor based on orientation tensors is proposed. This descriptor is obtained by combining polynomial coefficients calculated for each image in a video. The coefficients are found through the projection of the optical flow on Legendre polynomials, reducing the dimension of per frame motion estimations. The sequence of coefficients are then combined using orientation tensors. The global tensor descriptor created is evaluated by a classification of the KTH video database with a SVM classifier.

Combining Orientation Tensors for Human Action Recognition

This paper presents a new tensor motion descriptor based on histogram of oriented gradients. We model the temporal evolution of gradient distribution with orientation tensors in equally sized blocks throughout the video sequence. Subsequently, these blocks are concatenated to create the final descriptor. Using a SVM classifier, even without any bag-of-feature based approach, our method achieves recognition rates greater than those found by other HOG techniques on KTH dataset and a competitive recognition rate for UCF11 and Hollywood2 datasets.

Using graph cuts in GPUs for color based human skin segmentation

In this paper we propose a new method to deal with the problem of automatic human skin segmentation in RGB color space model. The problem is modeled as a minimum cost graph cut problem on a graph whose vertices represent the image color characteristics. Skin and non-skin elements are assigned by evaluating label costs of vertices associated to the weight edges of the graph. A novel approach based on an energy function defined in terms of a database of skin and non-skin tones is used to define the costs of the edges of the graph. Finally, the graph cut problem is solved in Graphics Processing Units (GPU) using the Compute Unified Device Architecture (CUDA) technology yielding very promising skin segmentation results for standard resolution video sequences. Our method was evaluated under several conditions, indicating when correct or incorrect results are generated. The overall experiments have shown that this automatic method is simple, efficient, and yields very reliable results.

A Tensor Motion Descriptor Based on Multiple Gradient Estimators

This work presents a novel approach for motion description in videos using multiple band-pass filters which act as first order derivative estimators. The filters response on each frame are coded into individual histograms of gradients to reduce their dimensionality. They are combined using orientation tensors. No local features are extracted and no learning is performed, i.e., the descriptor depends uniquely on the input video. Motion description can be enhanced even using multiple filters with similar or overlapping frequency response. For the problem of human action recognition using the KTH database, our descriptor achieved the recognition rate of 93.3% using three Daubechies filters, one extra filter designed to correlate them, two-fold protocol and a SVM classifier. It is superior to most global descriptor approaches and fairly comparable to the state-of-the-art methods.

A Video Tensor Self-descriptor Based on Block Matching

In this paper, we propose a new motion descriptor which uses only block matching vectors. This is a different and simple approach considering that most works on the field are based on the gradient of image intensities. The block matching method returns displacements vectors as a motion information. Our method computes this information to obtain orientation tensors and to generate the final descriptor. It is considered a self-descriptor, since it depends only on the input video. The global tensor descriptor is evaluated by a classification of KTH, UCF11 and Hollywood2 video datasets with a non-linear SVM classifier. Our results indicate that the method runs fast and has fairly competitive results compared to similar approaches. It is suitable when the time response is a major application issue. It also generates compact descriptors which are desirable to describe large datasets.

High Frequency Assessment from Multiresolution Analysis

We propose a method for the assessment and visualization of high frequency regions of a multiresolution image. We combine both orientation tensor and multiresolution analysis to give a scalar descriptor of high frequency regions. High values of this scalar space indicate regions having coincident detail vectors in multiple scales of a wavelet decomposition. This is useful for finding edges, textures, collinear structures and salient regions for computer vision methods. The image is decomposed into several scales using the Discrete Wavelet Transform (DWT). The resulting detail spaces form vectors indicating intensity variations which are combined using orientation tensors. A high frequency scalar descriptor is then obtained from the resulting tensor for each original image pixel. Our results show that this descriptor indicates areas having relevant intensity variation in multiple scales.

A Video Self-descriptor Based on Sparse Trajectory Clustering

In order to describe the main movement of the video a new motion descriptor is proposed in this work. We combine two methods for estimating the motion between frames: block matching and brightness gradient of image. In this work we use a variable size block matching algorithm to extract displacement vectors as a motion information. The cross product between the block matching vector and the gradient is used to obtain the displacement vectors. These vectors are computed in a frame sequence, obtaining the block trajectory which contains the temporal information. The block matching vectors are also used to cluster the sparse trajectories according to their shape. The proposed method computes this information to obtain orientation tensors and to generate the final descriptor. The global tensor descriptor is evaluated by classification of KTH, UCF11 and Hollywood2 video datasets with a non-linear SVM classifier. Results indicate that our sparse trajectories method is competitive in comparison to the well known dense trajectories approach, using orientation tensors, besides requiring less computational effort.

PanView: An Extensible Panoramic Video Viewer for the Web

In this paper we present a panoramic video viewer for the web. We propose Pan View, an open-source video-based panoramic viewer that provides a greater immersion of filmed environments. The main advantage of considering panoramic video is that users can pan or tilt virtual cameras in the recorded scene, allowing them to focus on information presented in different angles or moments, thus, allowing them to access more information than in ordinary videos or in static images. Pan View is fully extensible with easy customization using user coded modules and is implemented using modern web-browser standards, which reduces the computational requirements. To motivate the use of Pan View, we present a performance comparison considering a panoramic viewer based on Adobe Flash Player. The applicability of Pan View is shown in two projects: virtual tour on historical cities and analysis of a railroad network.

Discrete Wavelets on Edges

Human life is closely tied to signals. These signals are present everywhere listening to music is possible because of audible sound signals traveling through air, reading a book is feasible due to light waves bouncing off objects and interpreted by our bodies as visual images, electromagnetic waves allow us to communicate through the radio or wireless Internet. Signal Processing is an area of electrical engineering and applied mathematics that deals with either continuous or discrete signals. Particularly, Image Processing is any kind of Signal Processing where the input is an image, such as a digital photograph. The underlying essence of Image Processing lies in understanding the concept of what is an image and studying techniques for the manipulation of images with the use of a computer. While these explanations may seem quite generic, the importance of Image Processing in the modern world is undeniable and progress in this field is very desirable.