Long-Term Face Tracking for Crowded Video-Surveillance Scenarios

  title={Long-Term Face Tracking for Crowded Video-Surveillance Scenarios},
  author={Germ{\'a}n Barquero and Carles Fern{\'a}ndez Tena and I. Hupont},
  journal={2020 IEEE International Joint Conference on Biometrics (IJCB)},
Most current multi-object trackers focus on short-term tracking, and are based on deep and complex systems that do not operate in real-time, often making them impractical for video-surveillance. In this paper, we present a longterm multi-face tracking architecture conceived for working in crowded contexts, particularly unconstrained in terms of movement and occlusions, and where the face is often the only visible part of the person. Our system benefits from advances in the fields of face… Expand

Figures and Tables from this paper

Rank-based verification for long-term face tracking in crowded scenes
This paper presents a long-term, multi-face tracking architecture conceived for working in crowded contexts where faces are often the only visible part of a person, and demonstrates that this approach yields up to 50% longer tracks than state-of-the-art deep learning trackers. Expand


Adaptive appearance model tracking for still-to-video face recognition
Simulation results obtained with the Chokepoint video dataset indicate that the proposed method provides a significantly higher level of performance compared state-of-the-art systems when a single reference still per individual is available for matching. Expand
A Prior-Less Method for Multi-face Tracking in Unconstrained Videos
A prior-less method for tracking and clustering an unknown number of human faces and maintaining their individual identities in unconstrained videos is presented, which provides high accuracy in face association in the entire video with automatic cluster number generation, and is robust to outliers. Expand
MobiFace: A Novel Dataset for Mobile Face Tracking in the Wild
MobiFace, the first dataset for single face tracking in mobile situations, consists of 80 unedited live-streaming mobile videos captured by 70 different smartphone users in fully unconstrained environments and fine-tuning on the MobiFace training data significantly boosts the performance of deep learning-based trackers. Expand
Face-TLD: Tracking-Learning-Detection applied to faces
A novel system for long-term tracking of a human face in unconstrained videos is built on Tracking-Learning-Detection approach with the concept of a generic detector and a validator which is designed for real-time face tracking resistent to occlusions and appearance changes. Expand
Tracking Persons-of-Interest via Adaptive Discriminative Features
This paper extensively evaluates the proposed CNN-based multi-face tracking algorithm on a set of TV sitcoms and music videos and demonstrates significant performance improvement over existing techniques. Expand
Online multi-face detection and tracking using detector confidence and structured SVMs
FAST-DT integrates a generic face detector with an adaptive structured output SVM tracker and uses the detector's continuous confidence to solve the target creation and removal problem. Expand
An Automatic System for Unconstrained Video-Based Face Recognition
A robust and efficient system for unconstrained video-based face recognition, which is composed of modules for face/fiducial detection, face association, and face recognition is proposed. Expand
MOTS: Multi-Object Tracking and Segmentation
This paper creates dense pixel-level annotations for two existing tracking datasets using a semi-automatic annotation procedure, and proposes a new baseline method which jointly addresses detection, tracking, and segmentation with a single convolutional network. Expand
Distractor-aware Siamese Networks for Visual Object Tracking
This paper focuses on learning distractor-aware Siamese networks for accurate and long-term tracking, and extends the proposed approach for long- term tracking by introducing a simple yet effective local-to-global search region strategy. Expand
Fast Online Object Tracking and Segmentation: A Unifying Approach
This method improves the offline training procedure of popular fully-convolutional Siamese approaches for object tracking by augmenting their loss with a binary segmentation task, and operates online, producing class-agnostic object segmentation masks and rotated bounding boxes at 55 frames per second. Expand