Learning Sequential Visual Appearance Transformation for Online Multi-Object Tracking

Egileak: Itziar Sagastiberri Fernández Noud van de Gevel Jorge García Castaño Oihana Otaegui Madurga

Data: 18.11.2021


Abstract

Recent online multi-object tracking approaches combine single object trackers and affinity networks with the aim of capturing object motions and associating objects by using their appearance, respectively. Those affinity networks often build on complex feature representations (re-ID embeddings) or sophisticated scoring functions, whose objective is to match current detections with previous tracklets, known as short-term appearance information. However, drastic appearance changes during the object trajectory acquired by omnidirectional cameras causes a degradation of the performance since affinity networks ignore the variation of the long-term appearance information. In this paper, we deal with the appearance changes in a coherent way by proposing a novel affinity model which is able to predict the new visual appearance of an object by considering the long-term appearance information. Our affinity model includes a convolutional LSTM encoder-decoder architecture to learn the space-time appearance transformation metric between consecutive re-ID feature representations along the object trajectory. Experimental results show that it achieves promising performance on several multi-object tracking datasets containing omnidirectional cameras.

BIB_text

@Article {
title = {Learning Sequential Visual Appearance Transformation for Online Multi-Object Tracking},
pages = {176247},
keywds = {
Target tracking, Surveillance, Predictive models, omnidirectional Cameras, LSTM
}
abstract = {

Recent online multi-object tracking approaches combine single object trackers and affinity networks with the aim of capturing object motions and associating objects by using their appearance, respectively. Those affinity networks often build on complex feature representations (re-ID embeddings) or sophisticated scoring functions, whose objective is to match current detections with previous tracklets, known as short-term appearance information. However, drastic appearance changes during the object trajectory acquired by omnidirectional cameras causes a degradation of the performance since affinity networks ignore the variation of the long-term appearance information. In this paper, we deal with the appearance changes in a coherent way by proposing a novel affinity model which is able to predict the new visual appearance of an object by considering the long-term appearance information. Our affinity model includes a convolutional LSTM encoder-decoder architecture to learn the space-time appearance transformation metric between consecutive re-ID feature representations along the object trajectory. Experimental results show that it achieves promising performance on several multi-object tracking datasets containing omnidirectional cameras.


}
date = {2021-11-18},
}
Vicomtech

Gipuzkoako Zientzia eta Teknologia Parkea,
Mikeletegi Pasealekua 57,
20009 Donostia / San Sebastián (Espainia)

+(34) 943 309 230

Zorrotzaurreko Erribera 2, Deusto,
48014 Bilbo (Espainia)

close overlay

Jokaeraren araberako publizitateko cookieak beharrezkoak dira eduki hau kargatzeko

Onartu jokaeraren araberako publizitateko cookieak