AI Paper Atlas

Research map / Detection and segmentation / Tracking

Long-term Tracking in the Wild: a Benchmark

Jack Valmadre, Luca Bertinetto, Joao F. Henriques, Ran Tao, Andrea Vedaldi, Arnold W.M. Smeulders, Philip H.S. Torr, Efstratios Gavves

ECCV 2018 · Detection and segmentation › Tracking · approach: Transformer-based video object tracking · 159 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
add random re-detection logic with absent mode to SiamFCSiamFC · OxUvA testMaxGM: 0.313 -> 0.454Improved
add absent prediction via re-detection to SiamFCSiamFC · OxUvA testTNR: 0 -> 0.481Improved
add re-detection to SiamFCSiamFC · OxUvA testTPR: 0.391 -> 0.427Improved
add re-detection module to SiamFC versus post-hoc thresholdingSiamFC · OxUvA devTPR-TNR curveNo clear effect
apply post-hoc score thresholding to tracker scorestrackers with scores · OxUvA devTPR-TNR curveImproved
target disappearance in the videoall baseline trackers · OxUvA testTPRHurt
increase label frequency from 1Hz to 25Hz with fixed videos (ablation)OTB-100 evaluation · OTB-100AUC std deviation: between 1Hz and 25Hz no significant differenceNo clear effect
reduce label frequency to 0.5Hz (ablation)OTB-100 evaluation · OTB-100AUC std deviation: meaningful degradation at 0.5HzHurt
increase number of videos from 5 to 100 at 1Hz (ablation)OTB-100 evaluation · OTB-100AUC std deviation: steady and significant reduction in varianceImproved
annotate more videos sparsely (1Hz) rather than fewer videos densely (25Hz) (ablation)OTB-100 evaluation · OTB-100AUC std deviation: 4-5x smaller standard deviationImproved
longer tracking time beyond first minutemost trackers · OxUvA testTPRHurt
add 10% absent frames in the videoall trackers · OxUvA testTPR: 0% to 10% absentHurt

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 2 of 2 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy