Combining Local and Global Models for Robust Re-detection

Funding Agencies|SSF (SymbiCloud); VR (EMC2) [2016-05543]; CENIIT grant [18.14]; SNIC; WASP
Discriminative Correlation Filters (DCF) have demonstrated excellent performance for visual tracking. However, these methods still struggle in occlusion and out-of-view scenarios due to the absence of a re-detection component. While such a component requires global knowledge of the scene to ensure robust re-detection of the target, the standard DCF is only trained on the local target neighborhood. In this paper, we augment the state-of-the-art DCF tracking framework with a re-detection component based on a global appearance model. First, we introduce a tracking confidence measure to detect target loss. Next, we propose a hard negative mining strategy to extract background distractors samples, used for training the global model. Finally, we propose a robust re-detection strategy that combines the global and local appearance model predictions. We perform comprehensive experiments on the challenging UAV123 and LTB35 datasets. Our approach shows consistent improvements over the baseline tracker, setting a new state-of-the-art on both datasets.

Danelljan, Martin,1989-Linköpings universitet,Datorseende,Tekniska fakulteten(Swepub:liu)marda26 (author)
Khan, Fahad Shahbaz,1983-Linköpings universitet,Datorseende,Tekniska fakulteten,Incept Inst Artificial Intelligence, U Arab Emirates(Swepub:liu)fahkh30 (author)
Felsberg, Michael,1974-Linköpings universitet,Datorseende,Tekniska fakulteten(Swepub:liu)micfe03 (author)
Linköpings universitetDatorseende (creator_code:org_t)

In:Proceedings of AVSS 2018. 2018 IEEE International Conference on Advanced Video and Signal-based Surveillance, Auckland, New Zealand, 27-30 November 2018: Institute of Electrical and Electronics Engineers (IEEE), s. 25-30978153869294397815386929369781538692950

By the author/editor: Bhat, Goutam; Danelljan, Marti ...; Khan, Fahad Shah ...; Felsberg, Michae ...

About the subject

NATURAL SCIENCES: NATURAL SCIENCES; and Computer and Inf ...; and Computer Vision ...

Kungliga biblioteket hanterar dina personuppgifter i enlighet med EU:s dataskyddsförordning (2018), GDPR. Läs mer om hur det funkar här.
Så här hanterar KB dina uppgifter vid användning av denna tjänst.