Learning Sequential Visual Appearance Transformation for Online Multi-Object Tracking

Sagastiberri, Itziar; Gevel, Noud van de; García, Juan de Dios; Otaegui, Oihana

doi:10.1109/avss52988.2021.9663809

Cited by 3 publications

(8 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Both detection and tracking modules are the same as in our previous paper [1], specifically designed to track objects in videos recorded by omnidirectional cameras like the ones in this work. It relies on two main steps for each frame: (i) the detection of the object instances and (ii) the matching of detections to their corresponding tracklets.…”

Section: Selected Approachmentioning

confidence: 99%

“…All the clips were classified into three categories depending on the difficulty of the video stream: 1 • "Low": In these videos, there is usually one person who enters and/or exits the train door.…”

Section: Datasetmentioning

confidence: 99%

“…Our previous work explained how we obtain the best tracker configuration to achieve a state-of-the-art tracker [1].…”

Section: Experimental Evaluationmentioning

confidence: 99%

“…On the contrary, they do not work well with small groups of people, where an error of a single person can mean a significant change in accuracy. This paper is an extension of the work initially presented at the IEEE International Conference on Advanced Video and Signal-Based Surveillance [1]. In this case, the multi-object tracker analyses the people's trajectory to count the number of passengers inside a train.…”

Section: Introductionmentioning

confidence: 99%

“…Given this, the best-suited tracking method is the one that uses visual appearance to aid in the tracking, but it also needs to be robust against the distortion of the cameras. For the detection and tracking of people, the online multi-object tracking approach with an affinity model from previous work [1] is proposed.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Automatic Counting Passenger System Using Online Visual Appearance Multi-Object Tracking

Calle¹,

Sagastiberri²,

Aramburu³

et al. 2022

Adv. sci. technol. eng. syst. j.

View full text Add to dashboard Cite

In recent years, people-counting problems have increased in popularity, especially in crowded indoor spaces, e.g., public transport. In peak hours, trains move significant numbers of passengers, producing delays and inconveniences for their users. Therefore, analysing how people use public transport is essential to solving this problem. The current analysis estimates how many people are inside a train station by using the number of people entering and leaving the ticket gates or estimating the train occupancy based on conventional CCTV cameras. However, this information is insufficient for knowing the train occupancy. The required data includes vehicle usage: how many people enter or leave a vehicle or which door is the most used. This paper presents a solution to the stated problem based on a multi-object tracker with a sequential visual appearance predictor and a line-based counting strategy to analyse each passenger's trajectory using an overhead fisheye camera. The camera selection inside the train was made after profoundly studying the railway environment. The method proposes a module to compute the total train occupancy. The solution is robust against occlusions thanks to the selected tracker and the fisheye camera field of view. This work shows a proof of concept dataset containing pseudo-real case scenarios of people's affluence in train doors recorded by fisheye cameras. Its purpose is to prove the system's functionality in these scenarios. The proposed approach achieved an overall accuracy of counting people getting on and off of 90.78% in the pseudo-real dataset, proving that this approach is valid.

show abstract

Section: Selected Approachmentioning

confidence: 99%

Section: Datasetmentioning

confidence: 99%

“…Our previous work explained how we obtain the best tracker configuration to achieve a state-of-the-art tracker [1].…”

Section: Experimental Evaluationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Automatic Counting Passenger System Using Online Visual Appearance Multi-Object Tracking

Calle¹,

Sagastiberri²,

Aramburu³

et al. 2022

Adv. sci. technol. eng. syst. j.

View full text Add to dashboard Cite

show abstract

Deformable Modules for Flexible Feature Sampling on Vision Transformer

Park

Bahn

Jung

2022

2022 18th IEEE International Conference on Advanced Video and Signal Based Surveillance (AVSS)

View full text Add to dashboard Cite

A DC Grid-Connected PV Microgrid Regulated via Digital and MBPC Cascade Control Strategies

Gutiérrez¹,

Orellana²

2022

Adv. sci. technol. eng. syst. j.

View full text Add to dashboard Cite

The current paper focuses on the design of cascade control systems employed in photovoltaic (PV) powered microgrids (MG), which have been well received in remote agricultural farms in Ecuador. The study case is composed of a three-phase inverter powered by a DC-DC converter fed by a PV array and its control system. To simulate the effect of MG peak voltage regulation, a comparison of three cascade schemes was done, using MATLAB/SIMULINK®. These schemes are based on digital control, model-based predictive control (MBPC), and a combination of both. It was found that the best stabilization of voltage although with a fast response and a low overvoltage in the MG are reached with a combined scheme, compound by the classical controller combined with the MBPC. The results were validated using the RT-LAB software and the OPAL-RT real-time (RT) simulator, in the tested microgrid.

show abstract

Learning Sequential Visual Appearance Transformation for Online Multi-Object Tracking

Cited by 3 publications

References 27 publications

Automatic Counting Passenger System Using Online Visual Appearance Multi-Object Tracking

Automatic Counting Passenger System Using Online Visual Appearance Multi-Object Tracking

Deformable Modules for Flexible Feature Sampling on Vision Transformer

A DC Grid-Connected PV Microgrid Regulated via Digital and MBPC Cascade Control Strategies

Contact Info

Product

Resources

About