No abstract

Multiple Object Tracking (MOT) is the problem that involves following the trajectory of multiple objects in a sequence, generally a video. Pedestrians are among the most interesting subjects to track and recognize for many purposes such as surveillance, and safety. In recent years, Unmanned Aerial Vehicles (UAV’s) have been viewed as a viable option for monitoring public areas, as they provide a low-cost method of data collection while covering large and difficult-to-reach areas. In this paper, we present an online pedestrian tracking and re-identification framework based on learning a compact directional statistic distribution (von-Mises-Fisher distribution) for each person ID using a deep convolutional neural network. The distribution characteristics are trained to be invariant to clothes appearances and to transformations including rotation, translation, and background changes. Learning a vMF for each ID helps simultaneously in measuring the similarity between object instances and re-identifying the pedestrian’s ID. We experimentally validated our framework on standard publicly available dataset, which we used as a case study.

No abstract

In recent years, pedestrian re-identification has gained a lot of interest due to its importance for many purposes such as security and safety. Many types of solutions have been proposed to solve this problem, where the majority are based on a features extraction such Convolution Neural Networks (CNNs). These approaches assume that only the identities that are in the training data can be recognized. The pedestrians in the training data are called In distribution (ID). However, in real world scenarios, new pedestrians and objects can appear in the scene and the model should detect them as Out Of Distribution (OOD). In our previous study , we proposed a pedestrian re-identification based on von-Mises Fisher (vMF) distribution. Each identity is embedded in the unit sphere as a compact vMF distribution far from other identities distributions. The embedding is done through a base CNN. Recently, proposed a framework called Virtual Outlier Synthetic (VOS), that detects OOD based on synthesising virtual outlier in the embedding space in an online manner. Their approach assumes that the samples from the same object maps to a compact space. This assumption aligns with the vMF based approach. Therefore, in this paper, we revisited the vMF approach and merged with VOS in order to detect OOD data points. We present our approach to merge both frameworks. We conducted several experiments to evaluate our proposed framework. Results showed that our framework was able to detect new pedestrian that do not exist in the training data in the inference phase. It also slightly helped to improve the re-identification performance.

Pedestrian re-identification is an important field due to its applications in security and safety. Most current solutions for this problem use CNN-based feature extraction and assume that only the identities that are in the training data can be recognized. On the one hand, the pedestrians in the training data are called In-Distribution (ID). On the other hand, in real-world scenarios, new pedestrians and objects can appear in the scene, and the model should detect them as Out-Of-Distribution (OOD). In our previous study, we proposed a pedestrian re-identification based on von Mises–Fisher (vMF) distribution. Each identity is embedded in the unit sphere as a compact vMF distribution far from other identity distributions. Recently, a framework called Virtual Outlier Synthetic (VOS) was proposed, which detects OOD based on synthesizing virtual outliers in the embedding space in an online manner. Their approach assumes that the samples from the same object map to a compact space, which aligns with the vMF-based approach. Therefore, in this paper, we revisited the vMF approach and merged it with VOS to detect OOD data points. Experiment results showed that our framework was able to detect new pedestrians that do not exist in the training data in the inference phase. Furthermore, this framework improved the re-identification performance and holds a significant potential in real-world scenarios.