Abstract:Multi-object tracking (MOT) in the scenario of low-frame-rate videos is a promising solution to better meet the computing, storage, and transmitting bandwidth resource constraints of edge devices. Tracking with a low frame rate poses particular challenges in the association stage as objects in two successive frames typically exhibit much quicker variations in locations, velocities, appearances, and visibilities than those in normal frame rates. In this paper, we observe severe performance degeneration of many existing association strategies caused by such variations. Though optical-flow-based methods like CenterTrack can handle the large displacement to some extent due to their large receptive field, the temporally local nature makes them fail to give reliable displacement estimations of objects that newly appear in the current frame (i.e., not visible in the previous frame). To overcome the local nature of optical-flow-based methods, we propose an online tracking method by extending the CenterTrack architecture with a new head, named APP, to recognize unreliable displacement estimations. Further, to capture the fine-grained and private unreliability of each displacement estimation, we extend the binary APP predictions to displacement uncertainties. To this end, we reformulate the displacement estimation task via Bayesian deep learning tools. With APP predictions, we propose to conduct association in a multi-stage manner where vision cues or historical motion cues are leveraged in the corresponding stage. By rethinking the commonly used bipartite matching algorithms, we equip the proposed multi-stage association policy with a hybrid matching strategy conditioned on displacement uncertainties. Our method shows robustness in preserving identities in low-frame-rate video sequences. Experimental results on public datasets in various low-frame-rate settings demonstrate the advantages of the proposed method.

Spatial-temporal consistent labeling of tracked pedestrians across non-overlapping camera views

Tracking Multiple People under Occlusion and Across Cameras Using Probabilistic Models

Multiple Pedestrians Tracking Based on Gaussian Probability Model

A Novel Method For Real-Time Object Detection And Multiple Persons Tracking

APPTracker Plus : Displacement Uncertainty for Occlusion Handling in Low-Frame-Rate Multiple Object Tracking

Tracking human under occlusion based on kernel density estimation

Robust Visual Tracking Via CAMShift and Structural Local Sparse Appearance Model

Distributed Bayesian inference for consistent labeling of tracked objects in non-overlapping camera networks

Multi-person Multi-Camera Tracking for Live Stream Videos Based on Improved Motion Model and Matching Cascade

Retinoic Acid and Rapamycin Differentially Affect and Synergistically Promote the Ex Vivo Expansion of Natural Human T Regulatory Cells

Near-Online Multi-Pedestrian Tracking via Combining Multiple Consistent Appearance Cues

Multiple Human Association and Tracking from Egocentric and Complementary Top Views

Multihuman Tracking Based on a Spatial–Temporal Appearance Match

Multiview Pedestrian Localisation Via A Prime Candidate Chart Based On Occupancy Likelihoods

Joint Feature-Spatial-Measure Space: A New Approach to Highly Efficient Probabilistic Object Tracking

A Bayesian Detect to Track System for Robust Visual Object Tracking and Semi-Supervised Model Learning

Multi-CAMSHIFT for Multi-View Faces Tracking and Recognition

Multiple Human Tracking Based on Multi-view Upper-Body Detection and Discriminative Learning

Multiple-Target Tracking by Spatiotemporal Monte Carlo Markov Chain Data Association

Tracking-Assisted Object Detection with Event Cameras

Exploiting Pair-Wise Constraints Between Parts for Human Tracking