ssig_logo_bitmap_hd

The Smart Sense Laboratory, former Smart Surveillance Interest Group (SSIG), is coordinated by Prof. William Robson Schwartz and composed of researchers, graduate and undergraduate students that investigate problems related to Video Surveillance, Forensics and Biometrics by developing techniques on Computer Vision, Pattern Recognition and Digital Image Processing. The research group tackles problems including feature extraction, background subtraction, pedestrian detection, spoofing detection, person tracking, face recognition, person re-identification, pose estimation, gesture recognition, anomaly detection and activity recognition to allow scene understanding.

The laboratory provides a highly collaborative environment for its members by holding weekly meetings to discuss recent research papers, algorithms and research developments. In addition, the group maintains a computational infrastructure composed of powerful servers, large storage space, GPU cards and multiple surveillance cameras that allow the execution of large-scale experiments.

 

A new master among the group: Jessica Sena defends dissertation

A new master among the group: Jessica Sena defends dissertation

Jessica Sena defended this afternoon (18), her dissertation Human Activity Recognition based on Wearable Sensors using Multiscale DCNN Ensemble, obtaining the master degree from the Graduate Program in Computer Science of the Federal University of Minas Gerais (UFMG). Part of her research progress was presented about a month ago at the EUSIPCO conference in Rome, Italy, and Sena is ...
Read More
SIBGRAPI 2018 receive Victor Hugo Melo's presentation on activity recognition

SIBGRAPI 2018 receive Victor Hugo Melo’s presentation on activity recognition

Images extracted from the datasets used to train the method. Video understanding is the next frontier of computer vision, in which activity recognition plays a major role. The Ph.D. student Victor Hugo Melo will present his new approach to the subject, based on contextual cues obtained from object detections in the scene, in the main conference on computer vision ...
Read More
Active Cameras

Active Cameras

Unlike fixed cameras, statically positioned to capture a global view of the scene, active cameras or PTZs have a feature that allows them to move on the axis itself in the vertical, horizontal and zoom directions. In the scope of surveillance, PTZs are useful in viewing targets away from the fixed camera, such as people, car plates, abandoned objects in ...
Read More
Professor William Schwartz lectures on computer vision in video surveillance and biometry at UNESP

Professor William Schwartz lectures on computer vision in video surveillance and biometry at UNESP

This Tuesday (2), the amphitheater of the Faculty of Science and Technology of the Universidade Estadual Paulista (UNESP) hosts a lecture by Professor William Schwartz, head of Smart Sense Laboratory, in the 7th Week of the Course of Computer Science (SECOMPP), in the campus of Presidente Prudente. Closing the activities scheduled for the day, his speech will address the challenges of ...
Read More
/ News
Activity Recognition based on Wearable Sensors

Activity Recognition based on Wearable Sensors

Human activity recognition based on wearable sensors has received great attention in the past decade since it is fundamental to healthcare, homeland security and smart environments applications. In particular, human activity recognition based on wearable sensors has attracted the attention of the research community mainly due to easy acquisition and processing of the data. This task consists of assigning a ...
Read More
Gabriel Gonçalves will present his new approach for license plate recognition in SIBGRAPI 2018

Gabriel Gonçalves will present his new approach for license plate recognition in SIBGRAPI 2018

An essential task to ensure the identification of vehicles and safety on roads and private spaces, Automatic License Plate Recognition (ALPR) is the subject of research of Gabriel Gonçalves, Sense's Ph.D. student selected to present his progress in the Conference on Graphics, Patterns and Images (SIBGRAPI). In its 31st edition, SIBGRAPI is the main conference on computer vision in South ...
Read More
/ News
Jéssica Sena represents the Smart Sense Laboratory at the EUSIPCO conference in Italy

Jéssica Sena represents the Smart Sense Laboratory at the EUSIPCO conference in Italy

Last week, Rome was the venue for the 26th edition of the European Signal Processing Conference, EUSIPCO 2018, and Sense’s researcher Jéssica Sena was there to present her studies on Deep Convolution Neural Networks (DCNN) applied to human activity recognition based on wearable sensors. With Qualis A2 concept, the conference was held from 3 to 7 September and brought together ...
Read More
/ News
Researcher Rensso Colque gets Ph.D.'s degree in Computer Science

Researcher Rensso Colque gets Ph.D.’s degree in Computer Science

Rensso Mora Colque obtained this Friday (24) his Ph.D.'s title by the Program of Post-graduation in Computer Science of the Federal University of Minas Gerais (UFMG). His dissertation, entitled “Robust approaches for anomaly detection applied to video surveillance”, was guided by Sense head, professor William Robson Schwartz and presented three approaches to detect anomalous patterns in surveillance video sequences. A ...
Read More
/ News
Sense researchers make advances in controlling active cameras

Sense researchers make advances in controlling active cameras

                                              Unlike fixed cameras, statically positioned to capture a global view of the scene, active cameras or PTZs have a feature that allows them to move on the axis itself in the vertical, horizontal and zoom directions. In ...
Read More
/ Active Cameras, Machine Learning, News
SensorCap: Sensor Data Capture

SensorCap: Sensor Data Capture

SensoCap is an Android tool that captures sensor data in user-defined configurations. The purpose is to allow researchers and developers to quickly save sensor data for research, testing and prototyping. The sensors are broadly defined to include Motion, Position and Environment device sensors. An intuitive user interface makes tasks such as setting up a sensor, enter the capture details and ...
Read More
Benchmark on Activity Recognition based on Wearable Sensors

Benchmark on Activity Recognition based on Wearable Sensors

Source codes and data used in our paper Human Activity Recognition Based on Wearable Sensor Data: A Standardization of the State-of-the-Art. In this paper, we implement and evaluate several state-of-the-art approaches, ranging from handcrafted-based methods to convolutional neural networks. Also, we standardize a large number of datasets, which vary in terms of sampling rate, number of sensors, activities, and subjects. In ...
Read More
Job opportunity for developers on a project in parnership with Petrobras

Job opportunity for developers on a project in parnership with Petrobras

The SSIG is looking for developers to work on a Research and Development project in partnership with Petrobras S.A. The applicant will work on the development of a distributed, scalable smart video-surveillance system with real-time requirements using cutting-edge technologies to provide a high-performance system able to analyze the scene and provide a secure environment to the workers. We are hiring ...
Read More
/ News
SSIG offers masters scolarship on a project with Petrobras

SSIG offers masters scolarship on a project with Petrobras

The Smart Surveillance Interest Group offers a masters scholarship to work on the problem of active cameras in a large Research and Development project in partnership with Petrobras. The problem with Active Cameras is to control and coordinate a large set of cameras in order to capture visual data with a greater level of detail. For this, computer vision, machine ...
Read More
/ News
OFCM - Optical Flow Co-occurrence Matrices

OFCM – Optical Flow Co-occurrence Matrices

Source code of the spatiotemporal feature descriptor proposed in the Optical Flow Co-occurrence Matrices: A Novel Spatiotemporal Feature Descriptor (ICPR 2016). Aiming at capturing more information from the optical flow, this work proposes a novel spatiotemporal local feature descriptor called Optical Flow Co-occurrence Matrices (OFCM). The method is based on the extraction of Haralick features from co-occurrence matrices computed using the ...
Read More
Best Paper Runner-up Award - IJCB 2017

Best Paper Runner-up Award – IJCB 2017

SSIG paper Towards Open-Set Face Recognition using Hashing Functions received the Best Paper Runner-up Award in the International Joint Conference on Biometrics (IJCB 2017). This paper tackles open-set face identification. The problem of open-set face recognition determines whether the subject’s picture presented to the recognition system belongs to a known individual (person in the gallery). This problem is of great importance to find ...
Read More
/ Face Recognition, News
Importance Annotation for VIP and UIUC Pascal Sentence Datasets

Importance Annotation for VIP and UIUC Pascal Sentence Datasets

The experimental results for the paper Assigning Relative Importance to Scene Elements in SIBGRAPI’2017 (link to the research page) were obtained using two datasets: VIP dataset and UIUC Pascal Sentence. Both datasets are associate to importance assignment researches and present a wide range of images containing multiple objects per image. Data In order to use both datasets on the paper Assigning Relative Importance to Scene Elements, it ...
Read More
/ Datasets, Importance Assignment
Importance Assignment to Scene Elements

Importance Assignment to Scene Elements

The human brain is able to rapidly understand scenes through the recognition of their composing elements and comprehension of the role that each of them plays. This process, related to human perception, impacts in what people care when they see an image and the priority they give to each element. The idea of priority, also referred as importance, is based on biological features of ...
Read More
/ Current Research
R&D Project - Video Analytics Solutions

R&D Project – Video Analytics Solutions

The Smart Surveillance Interest Group started early this month the execution of a R&D project as a partnership with Maxtrack, a Minas Gerais based company leader in vehicle tracking and telemetry, through EMBRAPII DCC/UFMG. The project main focus is the development of algorithms based on computer vision and machine learning for video analytics considering Internet of Things (IoT) scenarios. Several problems in ...
Read More
/ News, Projects
SSAT - Smart Surveillance Annotation Tool

SSAT – Smart Surveillance Annotation Tool

Smart Surveillance Annotation Tool (SSAT), as it own name indicates, is an annotation tool, free and interactive, for the computer vision community. This eases the way researchers and annotation massive video datasets. This tool is still in progress, in the future it will be possible to create and manipulate bounding boxes in the video. Explanatory video Video extracted from Interactive Dataset, CPR contest ...
Read More
/ Annotation Tool, Software
2nd DeepEyes Workshop

2nd DeepEyes Workshop

The 2nd DeepEyes Workshop was held at UFMG Campus in October 03, 2016. Counting with around 70 people, including researchers such as Larry S. Davis (University of Maryland), François Bremond (INRIA), Eduardo Valle (Unicamp), Sandra Ávila, Filipe Costa, William Robson Schwartz (UFMG), were presented and discussed advances achieved in the previous year of the project in problems such as forgery detection on digital ...
Read More
/ News
SSIGLib - Smart Surveillance Interest Group Library

SSIGLib – Smart Surveillance Interest Group Library

The Smart Surveillance Interest Group Library (SSIGLib) is a C/C++ library built to provide a set of functionalities that aid researchers not only on the development of surveillance systems but also on the creation of novel solutions for problems related to video surveillance. The SSIGLib was designed to provide features for a good scene understanding, scalability, real-time operation, multi-sensor environment, ...
Read More
/ Software
Kernel Hierarchical PCA for Person Re-Identification (ICPR 2016)

Kernel Hierarchical PCA for Person Re-Identification (ICPR 2016)

Source code used in the Kernel Hierarchical PCA for Person Re-Identification (ICPR 2016). In this paper, we tackle the person re-identification problem as a common subspace learning and propose a novel framework, which we named Kernel HPCA, that handles with camera transition and dimensionality reduction. Kernel HPCA is a nonlinear extension of Hierarchical PCA (HPCA). HPCA is known in chemometrics as ...
Read More
AVSS2016 unsupervised prid450s

Kernel Partial Least Squares for Person Re-Identification (AVSS 2016)

Source code used in the Kernel Partial Least Squares for Person Re- Identification (AVSS 2016). In this paper, we approach supervised and unsupervised person re-identification (Re-ID) problem in two widely used datasets (VIPER and PRID450S) using Kernel Partial Least Squares. Regarding supervised Re-ID, we present a common subspace learning method, which we coined Kernel PLS Mode A, and a cross-view discriminative ...
Read More
SSIG License Plate Character Segmentation Database

SSIG License Plate Character Segmentation Database

This dataset, called SSIG SegPlate Database, aims at evaluating the License Plate Character Segmentation (LPCS) problem. The experimental results for the paper Benchmark for License Plate Character Segmentation  (link to the research page) were obtained using a dataset providing 101 on-track vehicles captured during the day. The video was recorded using a static camera in the early 2015. Data The images of the dataset were acquired ...
Read More
HOMF Descriptor for Anomalous Pattern Recognition

HOMF Descriptor for Anomalous Pattern Recognition

Source code used in the paper Histograms of Optical Flow Orientation and Magnitude to Detect Anomalous Events in Videos (published on SIBGRAPI 2015). In this paper, we propose the use of magnitude and orientation to describe patterns in crowded scenes. This model describes spatiotemporal regions in the scene to determine if they presents a normal or anomalous pattern.  Our descriptor ...
Read More
/ Feature Extraction, Software
Anomaly Detection Results

Anomaly Detection Results

This page shows results found in the literature for anomalous event detection in crowd data sets. If you like to have your published results added in the following tables, please send an e-mail to Rensso Mora with the link (or the pdf) to your paper and the results to be reported. Up to now, we have tabulated the results for the following datasets ...
Read More
/ Feature Extraction, Results, surveillance
Person Re-identification Results

Person Re-identification Results

This page shows results found in the literature for several person re-idenfication datasets (sort by rank-1 CMC). If you like to have your published results added in the following tables, please send an e-mail to Raphael Prates with the link (or the pdf) to your paper and the results to be reported. Up to now, we have tabulated the results for the following ...
Read More
/ Results
CBRA - Color-Based Ranking Aggregation for Person Re-Identification

CBRA – Color-Based Ranking Aggregation for Person Re-Identification

Source code used in the paper CBRA - Color-Based Ranking Aggregation for Person Re- Identification (ICIP 2015). In this paper, we propose the use of rank aggregation to improve the results, in which we address the person re-identification problem using a Color-based Ranking Aggregation (CBRA) method that explores different feature representations to obtain complementary ranking lists and combine them using the Stuart ranking ...
Read More
ICB 2015 Appearance-Based Person Re-Identification

ICB 2015 Appearance-Based Person Re-Identification

This page contains the Software used on the paper presented in ICB 2015 (Appearance-Based Person Re-Identification by Intra-Camera Discriminative Models and Ranking Aggregation). In this work, we used prototypes to indirectly handle with the camera transition problem. Thus, for a given gallery image, we computed the most similar individuals in training set that were captured by the same camera. For each ...
Read More
Smart Surveillance Framework

Smart Surveillance Framework

The Smart Surveillance Framework is a C/C++ library built using the OpenCV and the C++ Standard Template Library to provide a set of functionalities to aid researchers not only on the development of surveillance systems but also on the creation of novel solutions for problems related to video surveillance. One of its main goals is to provide a set of ...
Read More
/ Software, surveillance
PLS NIPALS C++

PLS NIPALS C++

This library provides a C++ class to execute Partial Least Squares (PLS) NIPALS method for a scalar response variable for both dimension reduction or regression. It provides a class composed of methods to build, load, and store a PLS model, project feature vectors onto the PLS model and retrieve its low dimensional representation. PLS handles data in high dimensional feature ...
Read More
/ Partial Least Squares, Software
Histogram of Shearlet Coefficients (HSC)

Histogram of Shearlet Coefficients (HSC)

This library provides a C++ class called HSC to perform feature extraction using the histogram of shearlet coefficients (HSC), method proposed in the paper A Novel Feature Descriptor Based on the Shearlet Transform (ICIP). Shearlet transforms provide a general framework for analyzing and representing data with anisotropic information at multiple scales. As a consequence, signal singularities, such as edges, can be ...
Read More
/ Feature Extraction, Software
PLS Face Identification

PLS Face Identification

This software, called PLS Face Identification (PFI), implements the one-against-all face identification method proposed in [1] and [2] (the second is an extension of the first by adding new feature descriptors). This software allows researchers to compare face identification methods to our method using datasets other than those considered in our papers. To facilitate its usage, we provide a small data ...
Read More
/ Face Recognition, Software
DetectorPLS

DetectorPLS

DetectorPLS is an implementation of the paper Human Detection Using Partial Least Squares Analysis.W.R. Schwartz, A. Kembhavi, D. Harwood, L. S. Davis. In proceedings of the ICCV. Kyoto, Japan, 2009 [pdf] [project webpage]. The goal of this implementation is to allow researchers to perform detection using already learned models for applications such as human and face detection and also provide a simple way ...
Read More
/ Pedestrian Detection, Software
ETHZ Dataset for Appearance-Based Modeling

ETHZ Dataset for Appearance-Based Modeling

The experimental results for the paper Learning Discriminative Appearance-Based Models Using Partial Least Squares in SIBGRAPI'2009 (link to the research page) were obtained using the ETHZ dataset, which provides a large number of different people captured in uncontrolled conditions. The video sequences are captured from moving cameras, which provides a range of variations in people's appearances. Data We used the ground truth location ...
Read More
/ Datasets, Person Re-identification