Model-based Multi-object Visual Tracking: Identification and Standard Model Limitations
This work addresses pedestrian tracking for surveillance or autonomous systems, but it is incremental as it adapts existing radar methods to a new domain without major breakthroughs.
The paper tackled pedestrian tracking using 2D bounding box detections by applying radar-based multi-object tracking methods, specifically the Poisson multi-Bernoulli mixture filter with a standard point-object model, revealing a model-data mismatch that limited performance.
This paper uses multi-object tracking methods known from the radar tracking community to address the problem of pedestrian tracking using 2D bounding box detections. The standard point-object (SPO) model is adopted, and the posterior density is computed using the Poisson multi-Bernoulli mixture (PMBM) filter. The selection of the model parameters rooted in continuous time is discussed, including the birth and survival probabilities. Some parameters are selected from the first principles, while others are identified from the data, which is, in this case, the publicly available MOT-17 dataset. Although the resulting PMBM algorithm yields promising results, a mismatch between the SPO model and the data is revealed. The model-based approach assumes that modifying the problematic components causing the SPO model-data mismatch will lead to better model-based algorithms in future developments.