AI Paper Atlas

Research map / Detection and segmentation / Open-set and few-shot detection

Robust Visual Tracking by Segmentation

Matthieu Paul, Martin Danelljan, Christoph Mayer, Luc Van Gool

ECCV 2022 · Detection and segmentation › Open-set and few-shot detection · approach: Amodal and open-world instance segmentation · 93 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 instance-branch conditioning of the segmentation decoder (ablation)RTS · LaSOTAUC: 65.3 -> 69.7Improved
add instance-branch conditioning (ablation)RTS · UAV123AUC: 65.2 -> 67.6Improved
add instance-branch conditioning (ablation)RTS · NFSAUC: 65.8 -> 65.4No clear effect
adapt LWL to tracking (RTS without instance branch)LWL · LaSOTAUC: 59.7 -> 65.3Improved
add instance-branch fallback when segmentation cannot localize target (ablation)RTS · UAV123AUC: 66.3 -> 67.6Improved
add instance-branch fallback (ablation)RTS · LaSOTAUC: 69.3 -> 69.7Improved
lower confidence threshold t_sc from 0.30 to 0.20 (ablation)RTS · LaSOTAUC: 69.7 -> 68.6Hurt
raise confidence threshold t_sc from 0.30 to 0.40 (ablation)RTS · NFSAUC: 65.4 -> 63.3Hurt
add auxiliary instance localization loss (eta=10) (ablation)RTS · LaSOTAUC: 67.7 -> 69.7Improved
increase eta from 0.4 to 10 (ablation)RTS · LaSOTAUC: 69.8 -> 69.7No clear effect
increase eta from 0.4 to 10 (ablation)RTS · GOT-10kAO: 84.0 -> 85.2Improved
fine-tune on YouTube-VOS with longer init and no scale change limitRTS · YouTube-VOS 2019G: 79.7 -> 80.3Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 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