AI Paper Atlas

Research map / Detection and segmentation / Tracking

STORM: Segment, Track, and Object Re-Localization from a Single Image

Yu Deng, Teng Cao, Hikaru Shindo, Quentin Delfosse, Jiahong Xue, Kristian Kersting

ICML 2026 · Detection and segmentation › Tracking · approach: Memory and correlation-based MOT · 0 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 attention (1 layer) to TOM matching (ablation)TOM · Tracking-loss verification (synthetic BOP)accuracy (%): 95.23 -> 98.36Improved
use learned TOM without attention instead of cosine similarity (ablation)TOM · Tracking-loss verification (synthetic BOP)accuracy (%): 87.55 -> 95.23Improved
stack a second attention layer in TOM (ablation)TOM · Tracking-loss verification (synthetic BOP)accuracy (%): 98.36 -> 97.84No clear effect
add TOM tracking-loss detection to FoundationPose tracking (ablation)STORM · Modified YCB-V tracking stress testADD AUC: 52.76 -> 74.64Improved
add TOM tracking-loss detection to FoundationPose tracking (ablation)STORM · Modified YCB-V tracking stress testAR: 50.09 -> 67.85Improved
add language conditioning to SOM (k=1) (ablation)SOM · BOP segmentation (k=1 reference views)AP_C: 67.3 -> 67.7Improved
add language conditioning to SOM (k=16) (ablation)SOM · BOP segmentation (k=16 reference views)AP_C: 67.9 -> 68.0Improved
increase reference views from 1 to 16 (with language) (ablation)SOM · BOP segmentationAP_C: 67.7 -> 68.0Improved
increase reference views from 1 to 16 (with language) (ablation)SOM · BOP segmentationlatency (ms): 45.8 -> 61.8Hurt
use 7 HSFA layers instead of 3 (ablation)SOM · SOM HSFA ablation test setAP (%): 1 percentage point higher AP (95.4%)No clear effect
use HSFA3 or HSFA7 instead of HSFA0 (simple attention, conv, linear) (ablation)SOM · SOM HSFA ablation test setAP (%): more than 50 percentage points higher than HSFA0Improved
align SAM3D reconstruction to benchmark coordinate frame (ablation)SAM3D + FoundationPose · YCB-V pose with SAM3D modelsVSD: 22.19 -> 60.72Improved

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