Object Tracking by Detection with Visual and Motion Cues

TL;DR

This paper introduces a novel online object tracking method combining visual appearance, motion cues, and a CNN-based metric estimation, evaluated on the BDD100K dataset with promising results.

Contribution

It proposes a new multi-frame CNN architecture for estimating tracking metrics and integrates these with a Kalman filter and assignment heuristic for improved tracking.

Findings

Multi-frame model achieves 39.1% MOTA on BDD100K

Single-frame model achieves 36.8% MOTA with low localization error

Efficient CNN architecture estimates appearance and motion metrics

Abstract

Self-driving cars and other autonomous vehicles need to detect and track objects in camera images. We present a simple online tracking algorithm that is based on a constant velocity motion model with a Kalman filter, and an assignment heuristic. The assignment heuristic relies on four metrics: An embedding vector that describes the appearance of objects and can be used to re-identify them, a displacement vector that describes the object movement between two consecutive video frames, the Mahalanobis distance between the Kalman filter states and the new detections, and a class distance. These metrics are combined with a linear SVM, and then the assignment problem is solved by the Hungarian algorithm. We also propose an efficient CNN architecture that estimates these metrics. Our multi-frame model accepts two consecutive video frames which are processed individually in the backbone, and then optical flow is estimated on the resulting feature maps. This allows the network heads to estimate the displacement vectors. We evaluate our approach on the challenging BDD100K tracking dataset. Our multi-frame model achieves a good MOTA value of 39.1% with low localization error of 0.206 in MOTP. Our fast single-frame model achieves an even lower localization error of 0.202 in MOTP, and a MOTA value of 36.8%.