The talk given by Laura Leal-Taixé at KUIS AI Talks on Oct. 21 in 2021.

Title: Shifting paradigms in multi-object tracking

Abstract:
The challenging task of multi-object tracking (MOT) requires simultaneous reasoning about track initialization, identity, and spatiotemporal trajectories. This problem has been traditionally addressed with the tracking-dy-detection paradigm. In this talk, I will discuss more recent paradigms, most notably, tracking-by-regression, and the rise of a new paradigm: tracking-by-attention. In this new paradigm, we formulate MOT as a frame-to-frame set prediction problem and introduce TrackFormer, an end-to-end MOT approach based on an encoder-decoder Transformer architecture. Our model achieves data association between frames via attention by evolving a set of track predictions through a video sequence. The Transformer decoder initializes new tracks from static object queries and autoregressively follows existing tracks in space and time with the new concept of identity preserving track queries. Both decoder query types benefit from self- and encoder-decoder attention to global frame-level features, thereby omitting any additional graph optimization and matching or modeling of motion and appearance. At the end of the talk, I also want to discuss some of our work in collecting data for tracking with data privacy in mind.

Short Bio:
Laura Leal-Taixé is a tenure-track professor at the Technical University of Munich, leading the Dynamic Vision and Learning group. She is a recipient of the Sofja Kovalevskaja Award from the Humboldt Foundation and the Google Faculty Award. She has been Area Chair for the major conferences in Computer Vision, was Program Chair of WACV last year, and will be General Chair of ECCV in 2024. Her research interests lie in dynamic scene understanding, visual localization, and object retrieval.

свернуть