NeighborTrack: Improving Single Object Tracking by Bipartite Matching with Neighbor Tracklets

We propose a post-processor, called NeighborTrack, that leverages neighbor information of the tracking target to validate and improve single-object tracking (SOT) results. It requires no additional data or retraining. Instead, it uses the confidence score predicted by the backbone SOT network to automatically derive neighbor information and then uses this information to improve the tracking results. When tracking an occluded target, its appearance features are untrustworthy. However, a general siamese network often cannot tell whether the tracked object is occluded by reading the confidence score alone, because it could be misled by neighbors with high confidence scores. Our proposed NeighborTrack takes advantage of unoccluded neighbors' information to reconfirm the tracking target and reduces false tracking when the target is occluded. It not only reduces the impact caused by occlusion, but also fixes tracking problems caused by object appearance changes. NeighborTrack is agnostic to SOT networks and post-processing methods. For the VOT challenge dataset commonly used in short-term object tracking, we improve three famous SOT networks, Ocean, TransT, and OSTrack, by an average of ${1.92\%}$ EAO and ${2.11\%}$ robustness. For the mid- and long-term tracking experiments based on OSTrack, we achieve state-of-the-art ${72.25\%}$ AUC on LaSOT and ${75.7\%}$ AO on GOT-10K.

PDF Abstract

Results from the Paper


Task Dataset Model Metric Name Metric Value Global Rank Result Benchmark
Visual Object Tracking GOT-10k NeighborTrack-OSTrack Average Overlap 75.7 # 1
Success Rate 0.5 85.72 # 2
Success Rate 0.75 73.3 # 1
Visual Object Tracking LaSOT NeighborTrack-OSTrack AUC 72.2 # 1
Normalized Precision 81.8 # 1
Precision 78.0 # 1
Visual Object Tracking TrackingNet NeighborTrack-OSTrack Precision 83.24 # 1
Normalized Precision 88.30 # 3
Accuracy 83.79 # 4
Visual Object Tracking UAV123 NeighborTrack-OSTrack AUC 0.725 # 1

Methods