atlas

Research map / Detection and segmentation / Tracking

Superpixel-Based Tracking-By-Segmentation Using Markov Chains

CVPR 2017 · Detection and segmentation › Tracking · approach: Foundation-model and correlation-filter tracking · 85 citations (OpenAlex)

Check your idea against 108,967 papers

AI-extracted summary. Written by AI from the paper's title only (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 SVR-learned scores to define edge weights instead of Lab-color distances (ablation)AMCT · NR, GBS, VS, ST2, DAVISaverage overlap ratio (segmentation masks): 23.1 -> 58.6; 53.0 -> 74.8; 71.4 -> 84.1; 47.2 -> 58.8; 41.2 -> 59.2Helped
Add a holistic target appearance model to select among connected-component candidates (ablation)AMCT · NR, GBS, VS, ST2, DAVISaverage overlap ratio (segmentation masks): 49.3 -> 58.6; 70.4 -> 74.8; 83.8 -> 84.1; 60.7 -> 58.8; 56.9 -> 59.2Helped
Add CNN feature descriptors to Lab color features for SVR (ablation)AMCT · NR, GBS, VS, ST2, DAVISaverage overlap ratio (segmentation masks): 58.6 -> 66.3; 74.8 -> 77.1; 84.1 -> 82.3; 58.8 -> 71.3; 59.2 -> 65.1Mixed
Use CNN features and ground-truth first-frame masks for online video segmentationAMCT+CNN* · NR, GBS, VS, ST2, DAVISaverage overlap ratio (online video segmentation): 66.4 vs 41.6; 79.0 vs 76.5; 89.7 vs 88.8; 74.5 vs 69.5; 73.2 vs 70.7Helped

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