AI Paper Atlas

Research map / Detection and segmentation / CNN features

IoUCert: Robustness Verification for Anchor-based Object Detectors

Benedikt Brückner, Alejandro J. Mercado, Yanghao Zhang, Panagiotis Kouvaros, Alessio Lomuscio

ECCV 2026 · Detection and segmentation › CNN features · approach: Box regression and loss design · 1 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
use AvgPool instead of MaxPool downsampling (ablation)IoUCert on YOLOv3-tiny (LARD 64x64) · LARD brightness eps=0.3robust cases (of 50): 15 -> 50Improved
use AvgPool instead of MaxPool downsampling (ablation)IoUCert on YOLOv3-tiny (LARD 64x64) · LARD brightness eps=0.3mean verification time (s): 1602.32 -> 56.26Improved
use AvgPool instead of MaxPool downsampling (ablation)IoUCert on YOLOv3-tiny (LARD 64x64) · LARD contrast eps=0.3robust cases (of 50): 6 -> 50Improved
use AvgPool instead of MaxPool downsampling (ablation)YOLOv3-tiny (LARD 64x64) · clean detectionmAP@0.5 (%): 86.88 -> 86.59No clear effect
use AvgPool instead of MaxPool downsampling (ablation)YOLOv3-tiny (LARD 64x64) · clean detectionmAP@0.5:0.95 (%): 41.67 -> 40.90Hurt
use tighter optimal IoU bounds instead of Cohen et al. bounds (ablation)IoUCert on SSD · SSD brightness eps=0.02 boundstightness improvement (%): 50.67 -> 65.09Improved
use tighter optimal IoU bounds instead of Cohen et al. bounds (ablation)IoUCert on SSD · SSD brightness eps=0.05mean verification time (s): 389.59 -> 404.54No clear effect
use tighter optimal IoU bounds instead of Cohen et al. bounds (ablation)IoUCert on SSD · SSD brightness eps=0.01branches explored: 74 -> 66Improved
use tighter optimal IoU bounds on shallow IoU ranges (ablation)IoUCert on SSD · SSD IoU bounds 0.50-0.60avoided exploration (%): 99.66Improved
increase perturbation budget eps from 0.01 to 0.30IoUCert on SSD · LARD brightnessrobust cases (of 50): 48 -> 9Hurt
increase input resolution from 64x64 to 128x128IoUCert on YOLOv3-tiny · LARD brightness eps=0.5robust cases (of 50): 47 -> 40Hurt
train on COCO instead of LARD at 128x128YOLOv3-tiny · brightness eps=0.5robust cases (of 50): 40 -> 29Hurt

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