Improved Person Re-Identification Using Statistical Approximation (original) (raw)
Related papers
Review of person re-identification techniques
IET Computer Vision, 2014
Person re-identification across different surveillance cameras with disjoint fields of view has become one of the most interesting and challenging subjects in the area of intelligent video surveillance. Although several methods have been developed and proposed, certain limitations and unresolved issues remain. In all of the existing re-identification approaches, feature vectors are extracted from segmented still images or video frames. Different similarity or dissimilarity measures have been applied to these vectors. Some methods have used simple constant metrics, whereas others have utilised models to obtain optimised metrics. Some have created models based on local colour or texture information, and others have built models based on the gait of people. In general, the main objective of all these approaches is to achieve a higher-accuracy rate and lowercomputational costs. This study summarises several developments in recent literature and discusses the various available methods used in person re-identification. Specifically, their advantages and disadvantages are mentioned and compared.
Fast person re-identification based on dissimilarity representations
Pattern Recognition Letters, 2012
Person re-identification is a recently introduced computer vision task that consists of recognising an individual who was previously observed over a video-surveillance camera network. Among the open problems, in this paper we focus on computational complexity. Despite its practical relevance, especially in real-time applications, this issue has been overlooked in the literature so far. In this paper, we address it by exploiting a framework we proposed in a previous work. It allows us to turn any person re-identification method, that uses multiple components and a body part subdivision model, into a dissimilarity-based one. Each individual is represented as a vector of dissimilarity values to a set of visual prototypes, that are drawn from the original non-dissimilarity representation. Experiments on two benchmark datasets provide evidence that a dissimilarity representation provides very fast re-identification methods. We also show that, even if the re-identification accuracy can be lower (especially when the number of candidates is low), the trade-off between processing time and accuracy can nevertheless be advantageous, in real-time application scenarios involving a human operator.
Different Techniques for Person Re-Identification: A Review
2019
In investigation, person re-identification is also a tough task of matching persons determined from utterly completely different camera views. It is necessary applications in AI, threat detection, human trailing and activity analysis. Person re-identification is also a tough analysis topic as a result of partial occlusions, low resolution images and massive illumination changes. Also, person determined from utterly completely different camera views has very important variations on poses and viewpoints. This paper summarises the challenges related to the person re-identification jointly discuss varied techniques utilized person re-identification.
Appearance Descriptors for Person Re-identification: a Comprehensive Review
In video-surveillance, person reidentification is the task of recognising whether an individual has already been observed over a network of cameras. Typically, this is achieved by exploiting the clothing appearance, as classical biometric traits like the face are impractical in real-world video surveillance scenarios. Clothing appearance is represented by means of low-level local and/or global features of the image, usually extracted according to some partbased body model to treat different body parts (e.g. torso and legs) independently. This paper provides a comprehensive review of current approaches to build appearance descriptors for person re-identification. The most relevant techniques are described in detail, and categorised according to the body models and features used. The aim of this work is to provide a structured body of knowledge and a starting point for researchers willing to conduct novel investigations on this challenging topic.
A Review on Person Re-Identification Techniques
2018
In investigation, person re-identification is also a tough task of matching persons determined from utterly completely different camera views. It is necessary applications in AI, threat detection, human trailing and activity analysis. Person re-identification is also a tough analysis topic as a result of partial occlusions, low resolution images and massive illumination changes. Also, person determined from utterly completely different camera views has very important variations on poses and viewpoints. This paper summarizes the challenges related to the person re-identification jointly discuss varied techniques utilized person re-identification. Index Terms Gaussian mixture model (GMM), Video-surveillance, Histogram of Oriented Gradients (HOG), Bidirectional ranking. ________________________________________________________________________________________________________
Dissimilarity-based people re-identification and search for intelligent video surveillance
"If opportunity doesn't knock, build a door." Anonymous Il Dottorato di Ricerca è una grande opportunità e insieme un gran rischio. Come il mio advisor (che ringrazierò opportunamente più avanti in queste righe) ama raccontare, è una "scatola vuota": robusta, ampia, ma pur sempre all'inizio vuota. E senza le istruzioni per riempirla. Il successo del tuo Dottorato dipende allora da cosa ci metti dentro. I tuoi colleghi, il tuo advisor, il tuo gruppo di ricerca, saranno fondamentali per dare forma e peso al contenuto, ma -parliamoci chiaro -l'onore e l'onere di riempirla sta a te, e solo a te.
PERSON RE-IDENTIFICATION USING MACHINE LEARNING
International Journal On Engineering Technology and Sciences – IJETS, 2024
Person Re-Identification (Re-ID) is a key computer vision problem that seeks to identify persons across non-overlapping camera images. This job is challenging due to the variations in lighting, location, and occlusion. In this publication, we review a variety of techniques and strategies applied in the Person Re-ID field. We discuss the evolution of Re-ID models, from manual feature-based methods to more advanced deep learning approaches. We also investigate the impact of model architecture, dataset size, and loss functions on the Re-ID system performance. With our experiments and discussions, we seek to shed light on the current state-of-the-art in Person Re-Identification and propose future directions of research for this exciting field.
Springer London eBooks, 2022
Networks of smart cameras share large amounts of data to accomplish tasks such as re-identification. We propose a feature selection method that minimizes the data needed to represent the appearance of objects by learning the most appropriate feature set for the task at hand (person re-identification). The computational cost for feature extraction and the cost for storing the feature descriptor are considered jointly with feature performance in order to select cost-effective good features. This selection allows us to improve inter-camera re-identification while reducing the bandwidth that is necessary to share data across the camera network. We also rank the selected features in the order of effectiveness for the task to enable a further reduction of the feature set by dropping the least effective features when application constraints require this adaptation. We compare the proposed approach with state-of-the-art methods on the i-LIDS and VIPeR datasets and show that the proposed approach considerably reduces network traffic due to inter-camera feature sharing while keeping the re-identification performance at an equivalent or better level compared with the state of the art.
Person re-identification by descriptive and discriminative classification
ABSTRACT Person re-identification, i.e., recognizing a single person across spatially disjoint cameras, is an important task in visual surveillance. Existing approaches either try to find a suitable description of the appearance or learn a discriminative model. Since these different representational strategies capture a large extent of complementary information we propose to combine both approaches. First, given a specific query, we rank all samples according to a feature-based similarity, where appearance is modeled by a set of region covariance descriptors. Next, a discriminative model is learned using boosting for feature selection, which provides a more specific classifier. The proposed approach is demonstrated on two datasets, where we show that the combination of a generic descriptive statistical model and a discriminatively learned feature-based model attains considerably better results than the individual models alone. In addition, we give a comparison to the state-of-the-art on a publicly available benchmark dataset.
A Framework for People Re-Identification in Multi-Camera Surveillance Systems
2017
People re-identification has been a very active research topic recently in computer vision. It is an important application in surveillance system with disjoint cameras. This paper is focused on the implementation of a human re-identification system. First the face of detected people is divided into three parts and some soft-biometric traits are extracted from each part. In second step, we can recognize people even if their faces are hidden or they are with back appearance. The features extraction will be carried out according to the overall characteristics of the complete images of different persons. An algorithm that identifies people from their body shape will be developed. A powerful representation of the person based on the characteristics of color, texture and shape as well as different soft-biometric features is suggested. The experiments are carried out on SAIVT-SoftBio database which consists of videos from disjoint surveillance cameras as well as some static image based dat...