ArTIST: Autoregressive Trajectory Inpainting and Scoring for Tracking
This work addresses the problem of reliable association in multiple object tracking for applications like surveillance and autonomous driving, presenting an incremental improvement by combining scoring and inpainting capabilities.
The paper tackles the challenge of scoring tracklet proposals in online multiple object tracking by introducing a probabilistic autoregressive generative model that measures the likelihood of natural motion, and demonstrates its effectiveness on MOT benchmark datasets while also showing generality in human motion prediction.
One of the core components in online multiple object tracking (MOT) frameworks is associating new detections with existing tracklets, typically done via a scoring function. Despite the great advances in MOT, designing a reliable scoring function remains a challenge. In this paper, we introduce a probabilistic autoregressive generative model to score tracklet proposals by directly measuring the likelihood that a tracklet represents natural motion. One key property of our model is its ability to generate multiple likely futures of a tracklet given partial observations. This allows us to not only score tracklets but also effectively maintain existing tracklets when the detector fails to detect some objects even for a long time, e.g., due to occlusion, by sampling trajectories so as to inpaint the gaps caused by misdetection. Our experiments demonstrate the effectiveness of our approach to scoring and inpainting tracklets on several MOT benchmark datasets. We additionally show the generality of our generative model by using it to produce future representations in the challenging task of human motion prediction.