![]() ![]() Alex Bewley, Zongyuan Ge, Lionel Ott, Fabio Ramos, and Ben Upcroft.EURASIP Journal on Image and Video Processing, Vol. Evaluating multiple object tracking performance: the clear mot metrics. Keni Bernardin and Rainer Stiefelhagen.In Proceedings of the IEEE/CVF International Conference on Computer Vision. Philipp Bergmann, Tim Meinhardt, and Laura Leal-Taixe.Delving deeper into convolutional networks for learning video representations. Nicolas Ballas, Li Yao, Chris Pal, and Aaron Courville.Experimental results on public datasets in various low-frame-rate settings demonstrate the advantages of the proposed method. Our method, with little additional computational overhead, shows robustness in preserving identities in low-frame-rate video sequences. Then we design a two-stage association policy where displacement estimations or historical motion cues are leveraged in the corresponding stage according to APP predictions. To overcome the local nature of optical-flow-based methods, we propose an online tracking method by extending the CenterTrack architecture with a new head, named APP, to recognize unreliable displacement estimations. Though optical-flow-based methods like CenterTrack can handle the large displacement to some extent due to their large receptive field, the temporally local nature makes them fail to give correct displacement estimations of objects whose visibility flip within adjacent frames. ![]() In this paper, we observe severe performance degeneration of many existing association strategies caused by such variations. Tracking with a low frame rate poses particular challenges in the association stage as objects in two successive frames typically exhibit much quicker variations in locations, velocities, appearances, and visibilities than those in normal frame rates. Multi-object tracking (MOT) in the scenario of low-frame-rate videos is a promising solution for deploying MOT methods on edge devices with limited computing, storage, power, and transmitting bandwidth. ![]()
0 Comments
Leave a Reply. |
AuthorWrite something about yourself. No need to be fancy, just an overview. ArchivesCategories |