Forward-Backward Error: Automatic Detection of Tracking Failures

By Zdenek Kalal, Krystian Mikolajczyk, Jiri Matas. 2010

914 citations recorded by the source corpus.

Subfields: Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Vision and Pattern Recognition

Citations in the corpus

Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks (2017, background): “In visual tracking, enforcing simple forward-backward consistency has been a standard trick for decades [24, 48].”
StrongSORT: Make DeepSORT Great Again (2022, background): “V-IOUTracker [5] extends IOUTracker [4] by falling back to single-object tracking [20, 25] while missing detection occurs.”
Near-Online Multi-target Tracking with Aggregated Local Flow Descriptor (2015, neutral): “(e.g. A1 = {d1, d10, d23}) during the time of presence; i.e. (V T1 ,DT1 )→ A.”
The Visual Object Tracking VOT2016 Challenge Results (2016, methodology): “These procedures are similar to the scale calculation of the median flow tracker [81], except that the clustering is used instead of median.”
Self-Paced Learning for Long-Term Tracking (2013, background): “Benchmark video comparison (F1 score) # Coke David Face1 Face2 Girl Moto-x Sylvester Tiger1 Tiger2 VW Panda Pedestrian Mean C1 LK-FB [29] 40.”
Supervision-by-Registration: An Unsupervised Approach to Improve the Precision of Facial Landmark Detectors (2018, methodology): “Therefore, the backward communication stream utilizes the forward-backward check [12] to evaluate the reliability of LK tracking.”
Recent advances and trends in visual tracking: A review (2011, background): “In Kalal et al. [111,112], proposed robust visual tracking algorithms based on the spatio-temporal constraints.”
Tracking-Learning-Detection (2012, methodology): “The tracking component of TLD is based on Median-Flow tracker [64] extended with failure detection.”
Learning Correspondence From the Cycle-Consistency of Time (2019, methodology): “Our work is influenced by the classic idea of forward-backward consistency in tracking [57, 79, 1, 27], which has long been used as an evaluation metric for tracking [27] as well as a measure of uncertainty [1].”
The Visual Object Tracking VOT2015 Challenge Results (2015, methodology): “These procedures are similar to the scale calculation of the median flow tracker [29], except that the clustering is used instead of median.”