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.

Partial Observation vs. Blind Tracking through Occlusion

Augmented tracking with incomplete observation and probabilistic reasoning

Robust Visual Tracking Via CAMShift and Structural Local Sparse Appearance Model

Tracking Multiple People under Occlusion and Across Cameras Using Probabilistic Models

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

Humans Tracking In The Complicated Background By Multi-Cue Integration

Visual tracking of partially observable targets with suboptimal filtering

An Occlusion Tolerent Method For Multi-Object Tracking

Tracking Occluded Objects And Recovering Incomplete Trajectories By Reasoning About Containment Relations And Human Actions

Partial Occlusion Handling for Visual Tracking Via Robust Part Matching

Tracking Multiple Objects Through Occlusion with Online Sampling and Position Estimation

Learning to Track Object Position through Occlusion

Object Tracking Using Spatio-Temporal Future Prediction

Multi-object Tracking Through Occlusions by Local Tracklets Filtering and Global Tracklets Association with Detection Responses

Standing Between Past and Future: Spatio-Temporal Modeling for Multi-Camera 3D Multi-Object Tracking

Multi-object Tracking by Expanding Long-Tracklets

Efficient Block-Division Model for Robust Multiple Object Tracking.

Visual Tracking Via Part-Based Model with Auxiliary Objects

Towards Occlusion Handling: Object Tracking with Background Estimation.

Robust and Accurate Object Tracking under Various Types of Occlusions

A Robust Part-Based Tracker