Flávio Luis Cardeal Pádua

Linear Sequence-to-Sequence Alignment

In this paper, we consider the problem of estimating the spatiotemporal alignment between N unsynchronized video sequences of the same dynamic 3D scene, captured from distinct viewpoints. Unlike most existing methods, which work for N = 2 and rely on a computationally intensive search in the space of temporal alignments, we present a novel approach that reduces the problem for general N to the robust estimation of a single line in RN. This line captures all temporal relations between the sequences and can be computed without any prior knowledge of these relations. Considering that the spatial alignment is captured by the parameters of fundamental matrices, an iterative algorithm is used to refine simultaneously the parameters representing the temporal and spatial relations between the sequences. Experimental results with real-world and synthetic sequences show that our method can accurately align the videos even when they have large misalignments (e.g., hundreds of frames), when the problem is seemingly ambiguous (e.g., scenes with roughly periodic motion), and when accurate manual alignment is difficult (e.g., due to slow-moving objects).

Linear sequence-to-sequence alignment

We present a novel approach for temporally aligning N unsynchronized sequences of a dynamic 3D scene, captured from distinct viewpoints. Unlike existing methods, which work for N = 2 and rely on a computationally-intensive search in the space of temporal alignments, we reduce the problem for general N to the robust estimation of a single line in RN. This line captures all temporal relations between the sequences and can be computed without any prior knowledge of these relations. Experimental results show that our method can accurately align sequences even when they have large mis-alignments (e.g., hundreds of frames), when the problem is seemingly ambiguous (e.g., scenes with roughly periodic motion), and when accurate manual alignment is difficult (e.g., due to slow-moving objects).

Temporal synchronization of non-overlapping videos using known object motion

This paper presents a robust technique for temporally aligning multiple video sequences that have no spatial overlap between their fields of view. It is assumed that (i) a moving target with known trajectory is viewed by all cameras at non-overlapping periods in time, (ii) the target trajectory is estimated with a limited error at a constant sampling rate, and (iii) the sequences are recorded by stationary cameras with constant frame rates and fixed intrinsic and extrinsic parameters. The proposed approach reduces the problem of synchronizing N non-overlapping sequences to the problem of robustly estimating a single line from a set of appropriately-generated points in R^N^+^1. This line describes all temporal relations between the N sequences and the moving target. Our technique can handle arbitrarily-large misalignments between the sequences and does not require any a priori information about their temporal relations. Experimental results with real-world and synthetic sequences demonstrate that our method can accurately align the videos.

SAPTE: A multimedia information system to support the discourse analysis and information retrieval of television programs

This paper presents a novel multimedia information system, called SAPTE, for supporting the discourse analysis and information retrieval of television programs from their corresponding video recordings. Unlike most common systems, SAPTE uses both content independent and dependent metadata, which are determined by the application of discourse analysis techniques as well as image and audio analysis methods. The proposed system was developed in partnership with the free-to-air Brazilian TV channel Rede Minas in an attempt to provide TV researchers with computational tools to assist their studies about this media universe. The system is based on the Matterhorn framework for managing video libraries, combining: (1) discourse analysis techniques for describing and indexing the videos, by considering aspects, such as, definitions of the subject of analysis, the nature of the speaker and the corpus of data resulting from the discourse; (2) a state of the art decoder software for large vocabulary continuous speech recognition, called Julius; (3) image and frequency domain techniques to compute visual signatures for the video recordings, containing color, shape and texture information; and (4) hashing and k-d tree methods for data indexing. The capabilities of SAPTE were successfully validated, as demonstrated by our experimental results, indicating that SAPTE is a promising computational tool for TV researchers.

A unified approach to content-based indexing and retrieval of digital videos from television archives

This work addresses the development of a unified approach to content-based indexing and retrieval of digital videos fromtelevision archives. The proposed approach has been designed to deal with arbitrary television genres, making it suitablefor various applications. To achieve this goal, the main steps of a content-based video retrieval system are addressed in thiswork, namely: video segmentation, key-frame extraction, content-based video indexing and the video retrieval operation itself.Video segmentation is addressed as a typical TV broadcast structuring problem, which consists in automatically determiningthe boundaries of each broadcasted program (like movies, news, among others) and inter-program (for instance, commercials).Specifically, to segment the videos, Electronic Program Guide (EPG) metadata is combined with the detection of two specialcues, namely, audio cuts (silence) and dark monochrome frames. On the other hand, a color histogram-based approach performskey-frame extraction. Video indexing and retrieval are accomplished by using hashing and k-d tree methods, while visualsignatures containing color, shape and texture information are estimated for the key-frames, by using image and frequencydomain techniques. Experimental results with the dataset of a multimedia information system especially developed for managingtelevision broadcast archives demonstrate that our approach works efficiently, retrieving videos in 0.16 seconds on average andachieving recall, precision and F1 measure values, as high as 0.76, 0.97 and 0.86 respectively.

Synchronizing Video Cameras with Non-overlapping Fields of View

This paper describes a method to estimate the temporal alignment between N unsynchronized video sequences captured by cameras with non-overlapping fields of view. The sequences are recorded by stationary video cameras, with fixed intrinsic and extrinsic parameters. The proposed approach reduces the problem of synchronizing N non-overlapping sequences to the robust estimation of a single line in RN+1. This line captures all temporal relations between the sequences and a moving sensor in the scene, whose locations in the world coordinate system may be estimated at a constant sampling rate. Experimental results with real-world sequences show that our method can accurately align the videos.

Evaluating cluster detection algorithms and feature extraction techniques in automatic classification of fish species

This paper proposes five different schemes for automatic classification of fish species. These schemes make the species recognition based on image sample analysis. Different techniques have been combined for building the classifiers: three feature extraction techniques (PCA, SIFT and SIFT + VLAD + PCA), three data clustering algorithms (aiNet, ARIA and k-means) and three input classifiers (k-NN, SIFT class. and k-means class) are tested. When compared to common methodologies, which are based on human observation, it is believed that these schemes are able to provide significant improvement in time and financial resources spent in classification. Two datasets have been considered: (1) a dataset with image samples of six fish species which are perfectly conserved in formaldehyde solution, and; (2) a dataset composed of images of four fish species in real-world conditions (in vivo). The five proposed schemes have been evaluated in both datasets, and a ranking for the methods has been derived for each one.

MIPS X-Ray: A plug-in to MARS simulator for datapath visualization

This paper presents the design and development of a new plug-in to the well-known MIPS Assembler and Runtime Simulator (MARS). The MIPS processor is a reduced instruction set computer (RISC), while the MARS simulator is a lightweight interactive development environment for programming in MIPS assembly language, intended for educational-level use. The proposed plug-in, called MIPS X-Ray, provides a dynamic dataflow diagram, which allows MARS users to visualize the execution of operations internally to the MIPS architecture. By using the proposed plug-in, MARS users can better analyze the developed assembly codes, improving their understanding about the MIPS processor, as well as their debugging capabilities.

A Local Feature Descriptor Based on Log-Gabor Filters for Keypoint Matching in Multispectral Images

This letter presents a new local feature descriptor for problems related to multispectral images. Most previous approaches are typically based on descriptors designed to work with images uniquely captured in the visible light spectrum. In contrast, this letter proposes a descriptor termed a multispectral feature descriptor (MFD) that is especially developed, such that it can be employed with image data acquired at different frequencies across the electromagnetic spectrum. The performance of the MFD is evaluated by using three data sets composed of images obtained in visible light and infrared spectra, and its performance is compared with those of state-of-the-art algorithms, such as edge-oriented histogram (EOH) and log-Gabor histogram descriptor (LGHD). The experimental results indicate that the computational efficiency of MFD exceeds those of EOH and LGHD, and that the precision and recall values of MFD are statistically comparable to the corresponding values of the forementioned algorithms.

Particle Filter-Based Predictive Tracking of Futsal Players from a Single Stationary Camera

In this paper we study the use of computer vision techniques for visual tracking of futsal players. In the sports field, player tracking is an important task, as it can provide an estimate of the position of the athlete in a given time and thus compute his/her trajectories. This information can be used by coaches and sport professionals on tactical and physical analyses. We use adaptive background subtraction and blob analysis to detect players, as well as particle filters to predict their positions and track them using data from a single stationary camera. Experimental results show that our approach is capable to track players and compute their trajectories over time with errors below 20 cm, thus demonstrating a high potential to be used in a wide range of futsal match analyses.