AI Paper Atlas

Research map / Detection and segmentation / CNN features

Normalizing Flow Based Feature Synthesis for Outlier-Aware Object Detection

Nishant Kumar, Siniša Šegvić, Abouzar Eslami, Stefan Gumhold

CVPR 2023 · Detection and segmentation › CNN features · approach: Region-based and query-based detectors · 13 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 a joint normalizing-flow distribution and synthesize low-likelihood outlier features instead of class-conditional Gaussian synthesisFFS object detector · PASCAL-VOC → MS-COCOFPR95: 47.77 -> 44.15Improved
Use a joint normalizing-flow distribution and synthesize low-likelihood outlier features instead of class-conditional Gaussian synthesisFFS object detector · PASCAL-VOC → OpenImagesAUROC: 87.59 -> 88.29Improved
Use flow-based feature synthesis rather than STUDFFS object detector · BDD100K → nuImagesmAP: 32.3 -> 36.2Improved
Use flow-based feature synthesis rather than STUDFFS object detector · Youtube-VIS → MS-COCOFPR95: 81.14 -> 83.06Hurt
Add rejection sampling to VOS outliers so samples unlikely under any class Gaussian are retained (ablation)VOS · PASCAL-VOC → MS-COCOFPR95: 48.22 -> 45.59Improved
Replace Glow with NICE (ablation)FFS · PASCAL-VOC → MS-COCOFPR95: 44.15 -> 48.23Hurt
Replace Glow with RealNVP (ablation)FFS · PASCAL-VOC → MS-COCOFPR95: 44.15 -> 45.76Hurt
Replace Glow with GIN (ablation)FFS · PASCAL-VOC → MS-COCOAUROC: 89.71 -> 88.36Hurt
Replace BCE regularization with CE (ablation)FFS · PASCAL-VOC → MS-COCOFPR95: 44.15 -> 69.92Hurt
Replace BCE regularization with JSD (ablation)FFS · PASCAL-VOC → MS-COCOFPR95: 44.15 -> 52.83Hurt
Replace BCE regularization with Hinge (ablation)FFS · PASCAL-VOC → MS-COCOFPR95: 44.15 -> 51.32Hurt
Increase the number of sampled projection outliers beyond a moderate amount (ablation)FFS projection sampling · PASCAL-VOC → MS-COCOFPR95: 47.23 -> 59.60Hurt

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