Tracking for Half an Hour
This addresses the challenge of stable long-term tracking for applications like surveillance or sports analysis, though it is incremental with improvements over existing methods.
The paper tackles the problem of long-term visual tracking, where existing trackers degrade over time due to model updates and target disappearance, and proposes a tracker that uses combined global and local search with cautious updating. It achieves state-of-the-art on the 100-sec UAV20L benchmark and outperforms others by a large margin on new half-hour videos.
Long-term tracking requires extreme stability to the multitude of model updates and robustness to the disappearance and loss of the target as such will inevitably happen. For motivation, we have taken 10 randomly selected OTB-sequences, doubled each by attaching a reversed version and repeated each double sequence 20 times. On most of these repetitive videos, the best current tracker performs worse on each loop. This illustrates the difference between optimization for short-term versus long-term tracking. In a long-term tracker a combined global and local search strategy is beneficial, allowing for recovery from failures and disappearance. Most importantly, the proposed tracker also employs cautious updating, guided by self-quality assessment. The proposed tracker is still among the best on the 20-sec OTB-videos while achieving state-of-the-art on the 100-sec UAV20L benchmark. On 10 new half-an-hour videos with city bicycling, sport games etc, the proposed tracker outperforms others by a large margin where the 2010 TLD tracker comes second.