Aljosa Osep

TL;DR: In this article, the authors extend the popular task of multi-object tracking to multiobject tracking and segmentation (MOTS) by creating dense pixel-level annotations for two existing tracking datasets using a semi-automatic annotation procedure, which includes 65,213 pixel masks for 977 distinct objects (cars and pedestrians) in 10,870 video frames.

TL;DR: This paper creates dense pixel-level annotations for two existing tracking datasets using a semi-automatic annotation procedure, and proposes a new baseline method which jointly addresses detection, tracking, and segmentation with a single convolutional network.

TL;DR: This work presents a novel MOT evaluation metric, higher order tracking accuracy (HOTA), which explicitly balances the effect of performing accurate detection, association and localization into a single unified metric for comparing trackers.

TL;DR: Higher order tracking accuracy (HOTA) as mentioned in this paper is proposed to explicitly balance the effect of performing accurate detection, association and localization into a single unified metric for comparing trackers, which is able to capture important aspects of MOT performance not previously taken into account by established metrics.

TL;DR: A differentiable proxy of MOTA and MOTP is proposed, which is combined in a loss function suitable for end-to-end training of deep multi-object trackers and establishes a new state of the art on the MOTChallenge benchmark.